Housing Policy:

Some Broader Perspectives

NEW ZEALAND BUSINESS ROUNDTABLE

JULY 1991

CONTENTS

EXECUTIVE SUMMARY i

1 INTRODUCTION 1

2 THE HOUSING MARKET IN CONTEXT 3

2.1 The Demand for Housing 3

2.2 The Supply of Housing 5

2.2.1 The Cost of Building Houses 5

2.2.2 The Supply of Housing to Owner-Occupiers 6

2.2.3 The Supply of Housing for Rent 7

2.3 Housing and the Labour Market 8

3 ECONOMIC AND REGULATORY POLICIES 9

3.1 Economic Management 9

3.1.1 Monetary Policy 9

3.1.2 Fiscal Policy 10

3.2 Finance 10

3.3 Taxation and Charges 12

3.3.1 Central Government Taxes and Charges 12

3.3.2 Local Authority Taxes and Charges 15

3.3.3 Directions for Reform 15

3.4 Planning and Development 17

3.5 Direct Building Costs 20

3.6 Labour Market 21

3.7 Rental Market 22

3.8 Housing Assistance 25

3.8.1 Economic Principles 25

3.8.2 Income Transfers, Vouchers and In-kind Assistance 28

3.9 Conclusion 31

4 DIRECT HOUSING ASSISTANCE 33

4.1 Introduction 33

4.2 General Lending 33

4.2.1 Housing Corporation 33

4.2.1.1 Modest-Income/Tenants Scheme 34

4.2.1.2 Homestart 37

4.2.1.3 Second Chance/Refinancing 39

4.2.1.4 Other Lending Schemes 40

4.2.2 Iwi Transition Agency 42

4.3 Insurance Schemes 42

4.3.1 Mortgage Guarantee 42

4.3.2 Building Performance Guarantee 43

4.4 Rental Housing 43

4.4.1 Housing Corporation 43

4.4.2 Iwi Transition Agency 50

4.4.3 Local Authority Housing 50

4.5 Department of Social Welfare -

Accommodation Benefit 50

5 EVALUATION OF DIRECT HOUSING ASSISTANCE 53

5.1 Introduction 53

5.2 Objectives 53

5.3 General Lending 54

5.3.1 Targeting 54

5.3.2 Level of Assistance 55

5.3.3 Incentive Effects 57

5.3.3.1 Labour Market 57

5.3.3.2 Tenure Choice 58

5.4 Mortgage Guarantee Scheme 58

5.4.1 Targeting 58

5.4.2 Level of Assistance 59

5.4.3 Incentive Effects 59

5.5 Rental Housing 60

5.5.1 Targeting 60

5.5.2 Level of Assistance 60

5.5.3 Incentive Effects 62

5.5.3.1 Labour Market 62

5.5.3.2 Rental Market 63

5.6 General Considerations 63

5.7 Directions for Reform 65

6 THE ROLE OF THE HOUSING CORPORATION 67

6.1 Introduction 67

6.2 Examination of State Businesses 67

6.3 The Housing Corporation's Role 68

6.4 Directions for Reform 68

7 CONCLUSIONS AND RECOMMENDATIONS 71

APPENDIX 75

I An Estimate of the Cost of Selected Forms of

Housing Assistance 75

II An Estimate of the Imputed Income Earned on

Owner-Occupied Houses 77

BIBLIOGRAPHY 79

ACKNOWLEDGEMENTS

The assistance of a large number of people with specialist knowledge of the housing industry and housing policy in contributing information for this study and reviewing draft material is gratefully acknowledged. Particular thanks are due to David Bower, Monier Brickmakers, Brian Conroy, New Zealand Housing Initiative, Dr Murray Coppen, formerly of United Bank, Keith Cullum, New Zealand Housing Industry Association, Ian Dickson, Fay Richwhite, Greg Dwyer, formerly of NZI Bank Limited, Lindsay Fergusson, Magnum Corporation, David Greig, The Treasury, John Hood, Fletcher Challenge, John Hunt, Building Industry Authority, Michael James, The Treasury, Susie Kerr, The Treasury, Graeme McIntosh, MGICA, Peter Menzies, Mainzeal Group, Graham Nakhies, formerly of the Housing Corporation, John Perera, Wellington Property Investors Association, Bill Roest, Fletcher Residential and Clive Thorpe, Trustbank. The aforementioned are not responsible in any way for the findings and views contained in the report.

Figures and information used in the report were current as of April 1991 unless otherwise stated.

EXECUTIVE SUMMARY

* Spending on housing services, consisting of homeownership and renting, comprises the single largest item of expenditure for the majority of New Zealand households. This fact, together with the pervasive role of the government in setting the rules within which the housing market operates and in providing housing assistance, motivated this broad review of the housing market and of the potential for improvements in policies which could enhance the welfare of the entire community.

* The government's management of the economy affects the market for housing services largely through the level of real and nominal interest rates and the degree of fluctuation and uncertainty in the market. More consistent and sustainable monetary and fiscal policies would reduce uncertainty and the risk component of interest rates.

* Following the deregulation of the financial sector, which commenced in 1984, competition among suppliers of home mortgages has increased, and funds have been allocated more efficiently. Further improvements in the quality of finance for housing purposes can be expected. The monopolisation by the Housing Corporation of the low-income, higher-risk segment of the market has constrained the choices available to such customers.

* The income tax system favours investment in homeownership relative to most other classes of investment. Under the goods and services tax, spending on housing services is treated more favourably than spending on other goods and services. These distortions can be expected to result in higher investment in owner-occupied housing and higher spending on housing services than would otherwise occur. Local authority taxes tend to discourage investment in housing, in land for housing purposes, or both. They may dampen the effects on the housing market of central government taxes.

* The cost of building houses in New Zealand is about 30-40 percent higher than in Australia. One of the principal reasons for this is the high cost of developing land to the point at which construction can start. It takes an average of between five and seven years to develop raw land to this point, most of which is taken in obtaining planning approvals.

* Building costs may also be unnecessarily high as a result of import protection, occupational licensing and excessive and prescriptive building standards. Continuing reductions in tariffs, and the removal of unjustified regulations which impede competition in key building trades, would help lower housing costs.

* Overall there appear to be major inefficiencies in the current planning and development process. There is considerable scope to make the existing system more flexible, to make greater use of market solutions to planning issues, and to reduce delays in obtaining approvals.

* Government interventions bias consumers in favour of homeownership and against renting. Such biases should be reduced over time in the interests of advancing consumer choice and improving efficiency. The development of a stronger private rental market is desirable, but it has been impeded by assistance to homeownership and by the Housing Corporation's rental housing programme.

* The Housing Corporation's lending and rental housing programmes have expanded substantially since 1984, and its rental stock is now at a record level.

* Many moderately well-off households and households with temporarily low incomes are assisted by the Housing Corporation. Much past lending does not appear to have been focused on households with serious housing needs.

* The main forms of housing subsidies are estimated to have amounted to around $1 billion in 1989/90. There is scope to reduce the total assistance to housing. The level of assistance per household varies considerably depending on particular circumstances.

* The Housing Corporation's lending and rental programmes can be expected to impose large efficiency costs on the community, particularly via the disincentive for individuals to earn income, reduced labour mobility and reduced choice of goods and services.

* Because it is not an efficient and fair means of achieving income distribution goals, the government should phase out most direct housing assistance. Consideration should also be given to reflecting (to a greater extent than is the case at present) differential housing costs in income maintenance programmes.

* If direct housing assistance is to be retained, it should be made neutral between homeownership and renting, and consumers should be given the choice of applying their assistance to mortgages or rental accommodation obtained from private sector suppliers.

* The Housing Corporation's business activities should be privatised. This should enable government debt to be reduced by over $6 billion.

* The reforms outlined would :

(i) improve consumer choice of housing;

(ii) reduce the cost of housing;

(iii) lead to a more economically efficient allocation of resources and hence a higher level of national output and income;

(iv) improve the government's fiscal position and significantly reduce its debt.

1. INTRODUCTION

The New Zealand Business Roundtable has published a number of studies on various aspects of New Zealand's welfare state aimed at promoting reforms which would:

(i) improve the quality, responsiveness and cost-effectiveness of social services. The real value of every dollar spent on social services, whether by individual New Zealanders directly or by the government on their behalf, could be increased by allowing greater freedom of choice in consumption and increased competition in delivery;

(ii) reduce the pressure of the welfare state on the government's budget. Lower government expenditure would help to reduce real interest rates, lower inflationary expectations and inflation, and lighten the tax burden. As a result, it would contribute to a sustained improvement in New Zealand's long-term economic growth rate and living standards;

(iii) increase the self-reliance of individual New Zealanders. Our society would be more harmonious if fewer people were trapped into dependence on the welfare state, and discouraged from building their lives with their own resources. As a result, the distribution of income in society could be a more equitable reflection of genuine need and effort.

Housing policy is an important part of the welfare state. Its impact on the government's financial position is large. It has the potential to trap many New Zealanders into debilitating states of dependence.

Moreover, it affects the consumption and delivery of what is regarded as one of the most important goods and services that society consumes. Planning procedures, building codes, tariffs, restrictive labour laws and tenancy regulations add to the cost of housing and reduce the choices open to consumers.

The purpose of this study is to put housing policy in context, to clarify how housing problems and policy responses can be analysed, and to identify policy reforms. It is organised into seven parts. After the introduction, the second part discusses the demand and supply factors affecting housing and examines the impact of housing on the labour market. The third discusses the main economic and regulatory policies affecting the housing market. The fourth and fifth parts discuss and evaluate respectively existing government housing programmes. The sixth examines the role of the Housing Corporation. The last part of the study presents our conclusions and recommendations.

Throughout the study, the focus is on housing services, i.e. accommodation in general. It is thus concerned with the accommodation provided by both rental housing and homeownership. The study does not, however, address housing issues concerning persons in institutional care, such as the mentally impaired or prison inmates. Nor does it examine tourist and holiday accommodation, such as motels and hotels, or accommodation provided free, e.g. by employers. These topics, while important, are not part of the central set of issues relating to the performance of the housing market.

2. THE HOUSING MARKET IN CONTEXT

To be seen in context, the housing market needs to be considered from three perspectives:

• that of consumers, i.e. occupants of dwellings, whose main concern can be expected to be the cost of purchasing housing services and the extent to which their particular housing preferences are satisfied. This is the demand side of the housing market.

• that of providers, i.e. landlords and owner-occupiers, whose main concern is the risk and return on their investment. This is the supply side of the housing market.

• that of the community as a whole, whose interest is in the efficiency of the housing market, including its impact on other markets, and in the extent to which housing opportunities and outcomes are equitable.

2.1 The Demand for Housing

According to the Household Expenditure and Income Survey 1989/90, New Zealanders spent 23 percent of their total expenditure on housing, either through renting or through meeting the costs of owning their own homes, or in buying and selling their houses.

Housing is on average the largest item of expenditure for New Zealanders. The performance of the housing market in delivering quality services at the lowest cost is therefore of interest to most people. The cost of housing also has an important bearing on the targeting of assistance to those in financial need.

New Zealanders appear to have been experiencing increasing costs of housing and/or an increasing preference for housing services relative to other forms of consumption. The increased rate of household formation and demand for greater quality may have been important factors in the latter case.

Demand for housing is affected by many factors including changing preferences regarding the size and quality of housing, demographic trends, government policy, nominal and real interest rates, income trends, the cost of housing services and other factors such as security and flexibility. These features are elaborated below.

Changes in the average size of households affect the demand for housing. In recent times, this has been falling in New Zealand. Whereas between 1981 and 1986 the total population increased by 4.1 percent, the number of households rose by 7.5 percent. This seems to have been caused mainly by two factors. First, the average age of the population has been increasing, which generally means smaller and more numerous households. Secondly, marital patterns have been changing; the proportion of single parent families, and people who never marry or are divorced or separated, has been increasing.

Whereas in the past most of the demand was for three-bedroom houses of standard design situated on quarter-acre sections, reflecting the predominance of the nuclear family, there is today a much greater diversity of demand. In particular, a greater proportion of housing consumers appears to want to live in smaller houses, with smaller sections and with more varied design features. This may be partly the result of changing household patterns, and partly due to changing lifestyle preferences.

A second factor affecting the demand for housing is the growth of the population, both throughout New Zealand and in particular areas. While the natural growth of the population in the country as a whole may be predicted with reasonable accuracy, the influence of external and internal migration flows is less certain. Net external migration is usually positive during periods of strong economic growth in New Zealand, and negative when the economy is in recession. Internal migration from small towns to the cities, and from the south to the north, although modified by local economic factors, has had a marked influence over time. The age structure of the population may also have an important influence on housing demand.

A third influence on housing demand is government policy, notably through the provision of below-market mortgages to owner-occupiers, and below-market rents to renters. In addition, government policy can influence expectations of future capital gains to be made from owning houses. Higher tax rates help to exacerbate the impact of capital gains which are not taxed.

Fourthly, nominal and real interest rates influence the overall demand for housing. The higher nominal interest rates are, the less potential purchasers can afford to borrow, given an unchanged income, to acquire an owner-occupied house or a rental property. Real interest rates impact on the demand for housing via the real return available on competing investments. A rise in real interest rates would tend to depress the demand for housing and reduce house prices relative to other prices in the economy.

Overall, between December 1974 and December 1987, the real price of housing services, both rental and owner-occupied accommodation, is estimated to have fallen. However, within this period, between December 1980 and December 1987, the real price of housing services rose by 29.5 percent.

Incomes are another influence on the demand for housing. As with most goods and services, people tend to spend more on housing as their incomes increase. However, in New Zealand, real disposable income fell between 1981 and 1988, while real expenditure on housing increased. This suggests that the income effect was offset by other factors, such as demographic influences, expectations of capital gains, and government policies.

Sixthly, the costs of housing services relative to those of other goods and services affect the overall demand for housing. For owner-occupiers, the cost of accommodation is affected by factors such as the price of the housing stock, expected capital gains, borrowing costs, operating costs and the required implicit return on the owner's investment. These factors are reflected in the assessed rental which the house would earn if it were let. This is known as imputed rent. For renters, the cost of accommodation primarily consists of the rent they pay.

As is discussed further below, when people decide to rent or buy, factors that are hard to value precisely also come into the picture. For example, owner-occupiers generally have greater security of tenure. On the other hand, renters generally have more flexibility.

Finally, the acquisition of a home reflects the decision both to acquire shelter and to invest in real estate. The ability to borrow against the security of real estate provides many households with a convenient means of building up their savings as the debt is subsequently repaid.

It is sometimes argued that people do not always make rational decisions in choosing whether to buy or rent accommodation. Commentators sometimes suggest that it would be economically more sensible for people to rent, yet most prefer to own their own homes. Similarly, it has been argued that being a landlord is economically irrational given market rents and an environment of low inflation. The implication of such observations is that public policy needs to take such 'irrationality' into account.

In evaluating those propositions, it needs to be recognised that it is not possible for observers to know the value individual homeowners and renters put on the choices open to them. Intangible factors associated with homeownership - such as security of tenure - may well add substantially to the overall welfare of many homeowners. If so, consumers may be prepared to accept quite high levels of monetary or 'economic' loss to retain these benefits. The fact that such intangible factors are difficult to value precisely - as are expectations of capital gains for many people - does not detract from the point that consumer decisions to rent or buy accommodation are implicitly based on a comparison of the imputed costs of ownership and renting. People know that they cannot receive capital gains or security of tenure from renting, or benefit substantially from making home improvements if they are renters, and these considerations rationally enter into their calculations, however intuitively they are made. Similar considerations apply to landlords.

To argue that most people (or agents acting on their behalf) are capable of making rational decisions about housing is not to imply that mistakes are not made - people can and do over-commit themselves, for example. It is simply to suggest that typically they do not systematically make decisions that are not in their own interest. Moreover, the capacity to make mistakes applies to choices in all other economic activities. The alternative to relying on individual decisions is to rely on decisions by third parties (for example by government officials). It does not seem convincing to argue that these are likely to be more rational since they cannot hope to be based on a knowledge of all the relevant circumstances that affect people's housing choices, nor on a strong set of incentives to make the right decisions.

2.2 The Supply of Housing

2.2.1 The Cost of Building Houses

The cost of building houses influences the value of existing houses. If a new house can be built for a lower cost than the value of equivalent existing houses, this will drive down existing house prices. Similarly, if existing house values are low relative to the cost of new houses, construction of houses will tend to slow until the demand for houses increases the value of existing houses. In these ways, the cost of new houses impacts on the value of all houses. The value of houses is a central factor in the determination of both explicit and imputed rents.

There are many factors influencing the cost of building houses. These include:

• the cost of developing the land and the opportunity cost of the land

• the cost of building materials

• costs relating to building codes

• the cost of labour, professional and trade services

• the cost of finance

• the overall relationship between supply and demand for houses. For example, the limited availability of developed sections may be a key constraint on supply responding to demand changes.

Regulations and practices affecting these costs are more closely examined in section 3. Reducing these costs may have a significant impact on overall economic efficiency as new investment in the housing stock comprises around 25 percent of all new investment.

2.2.2 The Supply of Housing to Owner-Occupiers

Owner-occupiers are in the position of being simultaneously renters and landlords. From the perspective of renters, they are concerned with their imputed rental costs. From the perspective of landlords, they are concerned with the value of their asset and its rate of return compared with alternative investments.

In 1986, there were about 1.08 million dwellings, of which 0.79 million (73 percent) were owner-occupied. The total value of owner-occupied dwellings in 1986 was about $67 billion, with the average value being around $85,300. The equity tied up in these dwellings was probably up to $60 billion, an average of up to $77,700. Around 43 percent of owner-occupied dwellings were not mortgaged.

While detailed information on wealth distribution in New Zealand is not available, there seems little doubt that, with the possible exception of the value of future superannuation entitlements, housing is the most important individual asset owned by most New Zealanders.

House prices are volatile. Between 1975 and 1988, the annual increase in house prices ranged from 2 percent to 30 percent. House prices rose twice as fast as inflation in 1974, between 1982 and 1984, and in 1988. On the other hand, they rose by less than inflation between 1976 and 1981, and in 1987.

Houses are illiquid assets, i.e. they cannot be sold quickly and cheaply for a constant relative value. The differences among houses, and changing consumer tastes (e.g. for location), mean that the value of houses relative to one another is not necessarily maintained in the period between purchase and sale. Moreover, selling and buying houses is a time consuming process, and the costs involved are high compared with buying or selling other assets like bonds and equities.

2.2.3 The Supply of Housing for Rent

The main factors which determine the cost of supplying rental accommodation are:

• the market value of the houses

• maintenance costs, rates and insurance

• the rate of return required by landlords, given that houses are risky assets. This includes the capital gains expected by them.

• taxation.

The cost of supplying rental accommodation and the demand for it interact to determine rents.

An increase in landlords' expected capital gains would tend to reduce the rents that they require. The changing value of houses may create capital gains and losses such that realised returns are different from expected returns. However, this does not in itself affect the assessment of the future cost of housing unless it changes expectations of future capital gains or losses.

If there were an increase in the rates of return on competing assets, for example yields on government bonds, rents would tend to increase. This process may well be slow as house investors take time to shift their investments away from houses to bonds or other assets. If the cost of constructing new houses increased, the market value of existing houses would rise and rents would also increase as a consequence.

As noted above, owner-occupiers implicitly take into account the same factors as landlords. Therefore, imputed rents would tend to rise and fall in much the same way as explicit rents. However, one key difference between owner-occupiers and landlords is that the former may have a more favourable taxation position, in that imputed rent is not taxed. It is worth noting that, if imputed rent were taxed, mortgage interest and other costs incurred by owner-occupiers should become tax-deductible, given the desirability of tax neutrality between owner-occupation and renting. These issues are elaborated in section 3.3.

One measure of the cost of renting is the ratio of rent to the average-male-ordinary-time-weekly wage. Between 1979 and 1988, it rose in Auckland from 31 percent to 61 percent, in Wellington from 31 percent to 50 percent, and in Christchurch from 22 percent to 42 percent.

Comparable figures for imputed rental costs are not available, mainly because expectations of capital gains are not directly observable. However, the ratio of mortgage repayments to average gross weekly earnings gives some indication of the cost of owner-occupied housing. For a 20 percent deposit on an average existing house purchased, weekly repayments as a proportion of weekly earnings rose between 1975 and 1986 from 43 percent to 59 percent.

2.3 Housing and the Labour Market

Housing is a major influence on the labour market because it affects the mobility of labour in a number of ways. First, the costs of buying and selling a house are considerable. It can easily cost people more to sell their house in one location and buy another house of the same quality elsewhere than the discounted annual increase in income they could gain by changing their place of work.

A related factor is that if people are unable to find suitable rental accommodation, and are forced to become owner-occupiers, they become less mobile than they would have preferred. Thirdly, if people are occupying subsidised rental accommodation, or are owner-occupiers with subsidised mortgages, it is likely that they are less mobile than if their housing were unsubsidised. It has been difficult to transfer subsidies.

The New Zealand economy is undergoing structural change, and can be expected to continue to do so as it becomes increasingly competitive. Labour mobility is crucial, to ensure both that workers are attracted to the jobs in which they are most productive and can generate maximum income, and that barriers to the unemployed seeking work in other locations are minimised.

A further issue is the impact on the incentive to work of housing subsidies. If unemployment and domestic purposes beneficiaries are able to gain preferential access to income-related housing subsidies, which are often very substantial, this can prove a major disincentive to their obtaining paid employment. Thus, one effect of housing subsidies can be to create dependency traps, reducing the number of people in paid employment and hence the growth of the economy.

3. ECONOMIC AND REGULATORY POLICIES

The foregoing discussion illustrates the underlying forces influencing the housing market and how they interrelate. The purpose of this section is to explore the main economic and regulatory policies affecting those forces, in particular the policies relating to:

• economic management

• finance

• taxation and charges

• planning and development

• building costs

• labour market

• rental market

• assistance to housing.

3.1 Economic Management

3.1.1 Monetary Policy

The aspect of economic management that probably most affects housing is monetary policy. In the long term, monetary policy appears to be the most important influence on inflation, even though in the short term other factors such as wage, exchange rate and interest rate movements and indirect tax changes may be more decisive.

If monetary policy is firm and consistent over a number of years, inflation can be expected to become low and stable. This should in turn cause expectations of future inflation to be low and more accurate, reducing the risk premium embedded in real interest rates.

In the past, the inflation risk premium embedded in New Zealand interest rates has probably been high by OECD standards. The reason for this seems obvious. While investors and borrowers are prepared to assign some probability to future inflation being within the government's target range of 0-2 percent by 1993, and staying there, they are also assigning a substantial probability to future inflation being outside this range.

There is every justification for this behaviour. In the recent past, New Zealand has not achieved a consensus on the importance of low inflation, and this has been reflected in the weak commitment of successive governments to disinflationary policies. The previous government experienced periods of vacillation, as did the present government when it was in opposition. The greater independence granted to the Reserve Bank to fight inflation and the enhanced accountability of the Reserve Bank to meet specific inflation targets may well turn out to be of critical importance in securing a continued commitment to price stability.

If New Zealand is successful in achieving 0-2 percent inflation by the end of 1993 or before, and in maintaining price stability thereafter, we can expect the credibility of monetary policy to increase, and consequently the inflation risk premium to reduce substantially.

This could have a major impact on housing. By reducing real financing costs, it would lower the cost of both consuming and supplying housing and the demand for housing would tend to increase as a result. It would tend to improve the living standards of lower income groups, thereby reducing the need to subsidise housing for this reason.

In addition, consistent monetary policy helps to ensure that strong swings in the provision and consumption of housing, and in the demand for inputs used in constructing and maintaining houses, do not occur. Such swings can cause sudden changes in housing prices, affecting owner-occupiers, renters and housing suppliers, thereby causing sudden wealth and income distribution changes. In consequence, they may cause additional economic uncertainty with its attendant costs, and misguided conclusions as to the ability of suppliers to respond to increases in demand.

3.1.2 Fiscal Policy

Swings in the government's financial surplus or deficit tend to induce fluctuations in economic activity, with a qualitatively similar undesirable effect on housing to that of swings in monetary policy.

In addition, fiscal policy is likely to have some impact on real interest rates because of its effect on the government's demand on the nation's savings. However, in an open economy, borrowers in New Zealand can call on the savings of foreigners. Hence, the impact of fiscal policy on interest rates is likely to be reflected most in the inflation risk premium, because of the impact of fiscal policy on inflation in the short term and also on the willingness of the government to maintain a firm monetary policy.

If the government is successful in eliminating its deficit, and in generating consistent surpluses for a period thereafter to reduce the stock of public debt, this would support a reduction in real interest rates. In addition, eliminating the strong swings in fiscal policy typical in the past in New Zealand could help to reduce the volatility in the supply of, and demand for, housing.

3.2 Finance

Since the deregulation of the financial services industry, creditworthy borrowers have had little difficulty in obtaining access to finance for housing as housing finance has been rationed mainly by price, rather than by regulatory constraints on lending. Hence, the dominant influence on the demand for housing finance is the level of nominal and real interest rates. However, there are a number of other aspects that are worth noting.

The first is that most mortgage finance is on a floating rate basis, with the interest rate being adjusted at the lender's discretion. Some lenders offer loans on which the rate of interest is fixed for at least part of the term of the loan. Competition among lending institutions generally ensures that individual institutions' lending rates do not move very far out of line with each other.

Margins on mortgages have been reducing over time, apparently for two main reasons. First, many lending institutions see mortgages as attractive assets on a risk and return basis, and have increased the share of mortgages in their portfolios. Secondly, the Reserve Bank requires registered banks to adhere to new capital adequacy ratio regulations. These require mortgages to be backed by only half the capital required for commercial loans. Whether this truly reflects the relative risks of such lending is open to serious doubt.

Home mortgage lenders have traditionally charged an up front fee of 1 percent, or more, of the amount of the loan to cover loan administration costs. Now, the level of the fee is generally negotiable, and is frequently waived, reduced or offset by a lower interest rate.

Competition has benefited people taking out mortgages in recent times and improvements in the market for housing finance are expected to continue. For example, further reductions in margins are likely to occur as average overhead costs are reduced. However, borrowers cannot be expected to be cross-subsidised. In the past, when interest rate controls were pervasive, savers found it difficult to earn interest on liquid funds. Now they can, and as a result banks have relatively small amounts of cheap deposits with which to cross-subsidise mortgages.

The mainstream private sector lending institutions may not fully meet the needs of the higher-risk end of the mortgage market. One reason for this may be the existence of the Housing Corporation. Because it is the channel by which most mortgage subsidies to homeowners with little equity and low incomes are provided by the state, it dominates this part of the market. As a result, the mainstream private sector institutions may have been crowded out. In addition, the Housing Corporation may have impaired the development of the private mortgage insurance market by offering a mortgage guarantee scheme for a nominal fee.

The Housing Corporation's position has meant in practice that prospective homeowners who are high risks or have low incomes have limited choice as to the institution from which they can borrow. One of the effects of this follows from the Corporation's reluctance to finance prospective homeowners into housing below a certain standard. While this may be justifiable from the Corporation's own perspective - it may wish for reputation reasons to have its lending associated with a certain standard of housing, or have its own views on what housing standards are appropriate - it may also have the consequence of limiting consumer choice.

Another example concerns Maori land. Until recently the Housing Corporation, the Iwi Transition Agency and its predecessor the Maori Affairs Department, would not lend unless the house providing the security was situated on freehold or leasehold land. This prevented some Maori from erecting houses on land which they owned communally. The Agency and the Corporation have changed this policy recently, with the result that some Maori have been able to live in more satisfactory circumstances than previously possible.

It should be noted that about one third of the current customers of the Corporation pay market rates on their mortgages, which may therefore readily be sold to private sector institutions. This would reduce the government's debt and business risk. The subsidised mortgages might also be able to be sold, particularly if the subsidy mechanism were changed.

3.3 Taxation and Charges

3.3.1 Central Government Taxes and Charges

The efficiency costs of raising government revenue are likely to be minimised if different classes of investment are taxed on a neutral basis. Where this condition is met, income tax does not alter the relative attractiveness to investors of different investment proposals. If certain classes of investment are favourably treated relative to others, then investment in these classes will be undertaken in place of those which are unfavourably treated. From a national viewpoint, output and incomes would, in all likelihood, be lower than otherwise.

Under one neutral form of income tax, the initial outlay (or capital expenditure) is taxed in the sense that it is financed out of tax-paid income and is not deductible, income earned during the life of the project is taxed on an accrual basis and the net cash flow paid out, including proceeds realised at the end of the project, is distributed free of tax. This treatment can be illustrated by reference to a savings account. The initial deposit is made out of tax-paid income, interest income is taxed as it is earned and the accumulated sum in the savings account can be withdrawn without incurring additional income tax. Such a regime has been described as a taxed/taxed/exempt regime and it has been applied to most asset classes. It is a standard which could be applied to all classes of investment provided that compliance and administration costs were not excessive.

In examining the appropriate income tax treatment of owner-occupied housing, it is helpful to view the owner as both an investor and a renter. The initial investment in an owner-occupied house is not deductible for income tax purposes and is therefore appropriately taxed in terms of the standard regime. The income earned by the owner, comprising imputed rents (net of current expenditure) is exempt from income tax whereas it would be taxable under the standard regime. Some homeowners are, however, able to arrange their financial affairs in such a way that a deduction may in effect be obtained for the cost of financing a house (and other assets which are treated on a similar basis, such as a boat). They may, for example, finance their houses from equity and employ a higher level of borrowed funds to finance other investments, the net income from which is taxable. In this case, a deduction for interest expense is effectively obtained where the income is exempt from tax. The final element of the tax regime, the proceeds realised on the sale of an owner-occupied house, is exempt from tax.

The income tax treatment of owner-occupied housing is concessional compared with the standard regime outlined above because net imputed rental income is not taxed. The fact that financing costs are effectively deducted in some cases increases this concession.

Given the high proportion of homeownership in New Zealand and the fact that households on higher incomes tend to own houses, it is probable that the exclusion of net imputed rental income from the tax base is a large component of government assistance to housing. Income tax forgone by not taxing imputed rents is estimated to have amounted to at least $450 million in 1989/90, assuming no behavioural responses to the imposition of such a tax. Details of this estimate are set out in Appendix II.

The income tax treatment of an investment in rental housing differs from that applicable to owner-occupied houses. The former is generally regarded as a taxable activity. As a consequence, the initial investment is not deductible, net rents are taxable and the proceeds realised on the sale of the property are generally free from tax. Up to 24 July 1990 any interest costs which had been deducted in respect of a rental property which was subsequently sold could be recovered if the property were sold within ten years of its purchase. Capital gains realised on rental properties are generally exempt from tax.

The following table summarises the tax treatment of rental and owner-occupied housing compared with a neutral regime.

Activity Income Tax Treatment of Housing

Owner- Neutral

Landlord Occupied Treatment

(1) Initial capital taxed taxed taxed

outlay

(2) Rent (or imputed taxed not taxed taxed rent)

(3) Current housing deductible not deductible deductible expenses

(e.g. interest)

(4) Proceeds from the not taxed not taxed not taxed

sale of the house

Note: Treatment of (2) means that owner-occupiers receive a tax concession, although if imputed rents were taxable then (3) would also logically become deductible for owner-occupiers.

The previous government's Consultative Document on the Taxation of Income from Capital, which was released on 19 December 1989, proposed that changes in the real value of owner-occupied houses (net of an allowance for maintenance) should be brought within the income tax net. The government announced at the time of its release that "family homes" would not be taxed. At that stage, the proposal would apparently have applied to other classes of homes, such as rental accommodation and holiday homes. In its March 1990 economic statement, the previous government announced that these proposals were "off the agenda". The present government has been opposed to introducing a capital gains tax.

The goods and services tax (GST) is intended to tax consumption spending. A consumption tax should generally apply to all items of consumption at a uniform rate as this is likely to minimise the efficiency cost of such a tax (e.g. the changes in consumption patterns induced by the imposition of the tax). In the case of housing, rents and imputed rents measure the consumption of housing services and should be included in the tax base provided that compliance costs are not excessive.

Rents and imputed rents on residential accommodation (other than short-term accommodation) are not subject to GST. Instead, GST is applied to the sale price of new houses, to services used in selling houses (e.g. real estate fees) and to inputs used in maintaining houses and in household operations (e.g. insurance). Sales of second-hand houses are exempt for GST purposes. As a consequence of these provisions, value added arising from the landlord's activities and the equivalent activities of an owner-occupier is not subject to GST.

A practical tax system requires a boundary to be drawn between taxable and non-taxable activities. Such a boundary can only be avoided if all activities, including leisure activities, are subject to tax. Once the boundary is drawn, there is an incentive to expand activities which are deemed to be non-taxable relative to those which are taxable and to organise activities in such a way that they are classified as non-taxable. Where the boundary between taxable and non-taxable activities is drawn is somewhat arbitrary and essentially depends on a judgment about the magnitude of the resulting distortions including the compliance and administrative costs involved.

It may be conjectured that the main reason for exempting homeowners from income tax on imputed rental income is the absence of arm's length transactions which would provide objective information on the amount of such income. Similar considerations apply in respect of GST. While it would be possible to levy GST on rents in respect of rental accommodation it would be more difficult to tax owner-occupiers on imputed rents. It would be necessary to register homeowners for GST purposes and to allow refunds of GST paid in respect of houses purchased and of maintenance and other costs. It is likely that this would bring up to a million additional entities into the GST collection system.

In the case of GST, owner-occupiers and renters have been treated on a uniform basis. Housing services as a whole are favoured relative to other goods and services to the extent that value added arises beyond the point at which GST is applied. In contrast, the income tax system favours homeownership over rental accommodation. This bias would have been accentuated if the proposal to tax capital gains in respect of homes other than family homes had been implemented.

Homeowners and landlords may be subject to two compulsory charges imposed by central government. First, if they insure their houses (or the contents thereof) for loss from fire, they are also required to insure such property for loss from earthquakes and similar perils. A levy of five cents for every $100 of cover is collected by insurance firms along with fire insurance premiums and paid (less a commission) to the Earthquake and War Damage Commission, a government-owned entity.

This method of insuring against natural disasters has been the subject of a review which started in 1988. A Bill was introduced into Parliament in which it was proposed that disaster insurance (on a replacement value basis) be made compulsory for most residential buildings. The Disaster Insurance Commission, a proposed successor to the present Commission, would have continued to have a monopoly over disaster insurance in respect of residential property. For the reasons set out in our submissions, which were prepared as a contribution to the review, it was argued that homeowners should be free to choose whether to insure their houses against natural disasters, that impediments to competition in the provision of disaster insurance should be removed and that the Earthquake and War Damage Commission should be subject to similar rules to those which apply to state-owned enterprises and should ultimately be privatised.

The second is the fire services levy which is also imposed if fire insurance is taken out. This levy finances the fire service. It is questionable whether the levy represents an optimal method of financing such services. Owners of uninsured property, for example, can benefit from fire services without contributing to the costs involved. Moreover, there are weak incentives for monitoring the efficiency and effectiveness of the fire service. For these reasons, we are of the opinion that the fire service levy should be reviewed.

3.3.2 Local Authority Taxes and Charges

Owner-occupiers and landlords are also required to pay taxes to local authorities. In 1989/90, rates (other than water rates) paid by owner-occupiers amounted to $674 million. There are three main bases on which territorial local authorities may calculate rates on residential properties. These are:

(i) the capital value system. Under this system the unimproved value of the land, together with the value of improvements (such as structures erected on the land), form the tax base. The capital value system is the most widely used one;

(ii) the land value system. The tax base in this case is the unimproved value of the land;

(iii) the annual value system. The annual rental value of the property forms the tax base.

Local authorities are permitted to charge different levels of rates on residential and non-residential properties. However, a Bill proposing the abolition of the annual value and land value rating systems and the use of differential rates (unless the differential can be justified on the basis of differing use of services) was introduced into Parliament prior to the 1990 general election and has been held over for consideration in 1991.

Local authority activities, other than trading operations (such as electricity distribution) and activities which are specifically subsidised by central government, are traditionally financed mainly by rates and borrowings. It is questionable whether rates are the most efficient form of finance for this purpose. From an economic point of view, the efficiency costs of local government activities would tend to be minimised if greater use were made of user charges and if taxes were broad-based. Property taxes are narrowly based, however, as they apply to one form of wealth and to one input used in providing housing services. As a result, homeowners are unlikely to use local authority services in proportion to their share of the rates burden.

The imputed net rental income of owner-occupiers in 1989/90 was estimated to be $1,370 million (see Appendix II). On this basis, local authority rates (other than water rates) of $674 million were broadly equivalent to an income tax of 49 percent. For the reasons set out in Appendix II, actual net income may be up to 1.6 times higher than this estimate, suggesting that the effective rate of tax is likely to be between 30 percent and 49 percent.

3.3.3 Directions for Reform

The economic consequences of the tax treatment of housing can be summarised as follows:

(i) investment in owner-occupied housing is preferred to most other investments, and hence it can be expected that investment in housing is higher than would otherwise be the case. The preferred treatment of housing for GST purposes can be expected to have the same result. The consumption of housing services can also be expected to be higher;

(ii) because the return on owner-occupied housing, adjusted for risk and after tax, can be expected to equal that of other investments, the distributional effects of the tax treatment are largely capitalised into house prices;

(iii) people on relatively high income tax rates will tend to own houses because they are best placed to benefit fully from the tax preference conferred on owner-occupied houses;

(iv) there is a bias toward homeownership at the expense of renting;

(v) local authority rating is likely to work in the opposite direction to the income tax system. The capital and rental value rating systems discourage investment in housing while the land value system discourages investment in land for housing purposes. Furthermore, the land value system can be expected to distort the investment in land relative to that in houses;

(vi) local authority rating encourages higher consumption of services associated with household operations, such as water and waste disposal, because there is only a weak link between the consumption of such services and the cost to individual homeowners.

As investment in housing is a large component of total investment and as housing expenditures are important elements of total consumer spending, it could be desirable to move toward a more neutral treatment of housing for tax purposes. This would improve the productivity of housing from a national point of view. While it might reduce the attractiveness of housing as an investment, the attractiveness of other investments (e.g. in fixed plant and equipment) would be correspondingly increased. Any such moves should desirably occur in the context of reductions in government spending and accompanying reductions in rates of tax, especially on income. Those moves would, in themselves, help to reduce distortions in the housing market. The set of reforms that could be relevant include the following:

(i) the imposition of income tax on the imputed rents of owner-occupiers in conjunction with (iv) below. Although there are data problems with this move, these may not be insurmountable. Imputed rent is taxed in Switzerland, for example, and New Zealand's income tax system already contains many imputed prices (e.g. depreciation allowances and livestock valuations). It may be possible to estimate approximate net rentals based on either market rents or the capital value of properties, or both;

(ii) the treatment of owner-occupied and other housing on a comparable basis to other classes of investment if any capital gains tax is implemented in future;

(iii) the reform of the Earthquake and War Damage scheme to enable homeowners and landlords to choose whether to insure against loss from natural disasters and to permit a competitive market in the provision of disaster insurance to develop;

(iv) the reform of local authority financing by extending user charges where feasible and by examining the desirability of financing other services via a broader tax than rates.

3.4 Planning and Development

Discussions with people in the industry indicate that it is around 30-40 percent more expensive to construct a new house in New Zealand than in Australia. One of the principal reasons for this is the high cost in New Zealand of developing land to the point at which house construction can begin.

This is partly due to the artificial scarcity of land zoned for housing at any one time. It is also due to the delays which involve high interest charges while the land is held. A further factor is the additional costs imposed directly by local authorities.

The artificial scarcity of land is caused partly by the zoning system. This limits the amount of land that may be used for housing. As a result, the cost of land zoned for housing, especially for higher density housing, is relatively expensive.

The delays are twofold. First, there are delays in getting raw land to the stage of being properly zoned for housing in district schemes, or of being usable for housing through specified departures from the schemes. Secondly, there are delays in developing the land from this point to the stage at which it can be built on.

Indications from builders and developers in the industry are that it takes an average of between five and seven years for unzoned raw land to reach the point at which housing can be built on it. Much of this time can be spent advocating and waiting for zoning changes or specified departures.

About two years is taken up by the process of developing zoned raw land into sections which can be built on. This includes designing and making roads, providing services such as drainage and storm water, and defining individual titles to the properties. The physical work takes on average between four and six months; the rest of the time is spent negotiating with local authorities and waiting for them to approve applications.

Local authorities must approve concept plans, scheme plans and engineering plans, and this typically takes eighteen months. It is also common for local authorities to impose conditions at the last minute, leaving developers with no choice but to accept them because of the high costs of further delay.

Local authorities impose costs in other ways. First, the cost of obtaining permits, such as building permits, is high relative to the work involved. Secondly, the development standards they impose through their ordinances in residential areas may be more restrictive than necessary. For example, the size and quality of streets and footpaths are designed to minimise maintenance costs, which are borne by local authorities, rather than to meet the realistic needs of the community. The amount of off-street and on-street parking, the distance houses must be from the borders of sections, and the size of drains are further instances of higher standards than are probably necessary in many situations.

A further problem is the compulsory revenue contribution, typically 7.5 percent of the total cost of the development, and the fact that some local authorities take it in cash and appear to treat it as a form of income.

Local authority ordinances not only add substantially to the cost of supplying land for housing; they may also reduce its quality. They may be suitable for subdivisions for three bedroom houses for nuclear families but are often unsuitable for the housing required by the increasing proportion of smaller households, and the greater diversity of lifestyles and housing preferences.

Over the last few decades, the situation has deteriorated in many ways. Indications from builders and developers in the industry are that land costs used to be on average around 20 percent of the total cost of a new house, but they are now around 30 percent. The New Zealand planning process has become highly regulated and legalistic. It has on the surface become more democratic, by apparently allowing more public participation, but in reality it has disenfranchised people because the additional costs have limited their housing choices.

It is symptomatic that builders now seek to minimise the amount of planning risk they take by buying fully developed land or undertaking only minimal development. Zoning risk is no longer taken by major firms as a general rule.

Moreover, the situation may get worse. The amalgamation of local authorities that has recently taken place may be an unfavourable development in this respect. The larger authorities in the past have tended to be the ones causing the longest delays in the development process, whereas some of the smaller authorities have been more responsive and flexible. On the other hand some of the new authorities appear to be willing to adopt a more flexible procedure.

Secondly, there is concern that the Resource Management Act will also make matters worse. While in theory it might improve the process by providing a 'one stop shop' for obtaining planning consents, and by setting time limits for various parts of the process, a 'one stop shop' is illusory when developers have to deal separately with different departments of local authorities, and statutory time limits do not work at present.

Of greater concern is that the Act removes constraints upon local authorities' ability to impose conditions on developments, and allows anyone to object to developments, even if they have no direct interest. Moreover, it may take 5 to 10 years before sufficient precedents exist to give developers certainty as to how the new Act will be interpreted; at the moment, the forty years of Planning Tribunal decisions provide a fair degree of certainty on this question. The first 2 to 3 years after the passage of the Act can be expected to be very difficult for developers.

The Act does little to encourage the development of voluntary agreements as an alternative to regulation. There is an international trend towards using market solutions, and this is starting to become evident in New Zealand. It is manifest in developers negotiating with local authorities to compensate the community for modifying planning norms by providing other amenities. However, its potential for avoiding the artificial scarcities caused by regulation, the lengthy delays due to the cumbersome process for changing permitted land uses, and the often unfair and arbitrary outcomes caused by politicised procedures, has hardly been tapped. Compensation to individuals for losses imposed by planning decisions has rarely been made in New Zealand.

In general, regulation is necessary and efficient only where property rights are not sufficiently well specified. A land use planning process, including hearings and quasi-judicial procedures of various kinds, is only necessary where voluntary agreement is not an efficient mechanism for reaching solutions.

However, there is considerable scope to increase the extent to which property rights are defined and transferred in New Zealand, for example by covenants attached to property titles which can be modified by negotiation or voting.

The private street phenomenon is an example of residents voluntarily agreeing to change the nature of their property rights, and attaching covenants to their titles, as a means of improving their living environment.

There are a number of private streets that already exist which generally take the form of self contained precincts. They are usually the result of a developer's inability to build to the density required with individual freehold titles, owing to the constraints of local planning ordinances. The subsequent development is achieved by resorting to the device of unit title or crosslease. In either case the local authority will not maintain roading or services within the complex and mail delivery and rubbish collection takes place at the entrance from the public road. Though most of the unit title developments are satisfactory, those on crosslease and directed at the bottom of the market have few amenities and these are unlikely to be improved or properly maintained as the co-owners lack the financial means.

In private streets, aspects of the property rights affecting many residents may be changed following a voting procedure. The reason for this is that it may be difficult for pure market solutions to be reached, because of the number of people that are involved in the negotiation process and the incentives that may exist for individuals to stymie negotiations in order to extract excessive compensation.

This illustrates the point that there is no absolute choice between regulation and planning processes on the one hand, and private property rights and voluntary agreements on the other. It is more a question of allowing and encouraging voluntary processes to have greater scope, with regulation and quasi-judicial processes serving as a backstop to private processes.

Regional and local authorities have a key role to play in this, because people will not put any effort into seeking voluntary agreements to reallocate property rights if they think that the authorities will frustrate them. Moreover, the authorities have to be a party to most such agreements because of the implications for the services they provide. For example, there is a disincentive for residents to form a private street and take over responsibility for its maintenance to the standard they desire if there is not a corresponding reduction in their rates.

An important recent development is the provision in the Resource Management Act which requires an explicit evaluation of the economic costs and benefits of any regulatory intervention, and for the 'do nothing' option to be considered. If adopted and applied rigorously, this could significantly cut down the number of harmful interventions.

The Housing Initiative, an incorporated society with open membership but currently receiving strong support from the Housing Corporation and leading firms in the building industry, has been strongly urging local authorities to allow changes in sub-divisional practices so as to permit the better use of scarce land resources through higher density developments where each unit has freehold status. The concepts of the Initiative are also aimed at reducing ordinance-generated costs so that the savings made can be employed in improving the quality of housing, the range of choice and the living environment.

Many local authorities are currently considering scheme changes in line with Initiative concepts, and some have already made the adjustment. The influence of the Initiative has also had a significant impact on the thinking of the industry and its servicing professional groups, and many of its concepts have already successfully been applied in public and private housing developments.

There is considerable further scope to lower housing costs by improving the regulation of planning and development. Desirable moves would include:

• the more widespread adoption by local authorities of Housing Initiative principles, to lower the cost and improve the quality of housing for consumers

• the streamlining of planning processes to reduce delays and to narrow and define the scope for objections

• the use of competition and incentives for planners and inspectors to carry out their duties in the most efficient fashion

• the introduction of procedures which would allow greater definition and transferability of private property rights in the planning and development process.

3.5 Direct Building Costs

There are three main elements of direct building costs which affect the price of housing. These are the cost of building materials, the cost of the building process and the effect of the building code.

During recent years, the removal of import licensing and the reduction of tariffs has had a beneficial impact on the cost of building materials. They have also forced domestic suppliers to become more efficient, and engendered greater competition among them by destroying cosy market sharing.

There is scope to reduce building material costs further where the costs of transport and tariffs remain significant. For example, bricks in New Zealand are significantly more expensive than they are in Australia, and many fittings are as much as 25 percent cheaper on the other side of the Tasman. Cement is also relatively expensive in New Zealand, apparently because of high shipping costs although specific figures were not available to substantiate this. Indirect costs to the building industry, such as the cost of motor vehicles which is higher because of tariff protection, are also important. The recent decisions to liberalise coastal shipping and the government's stated intention to permit competition in trans-Tasman shipping may be helpful in reducing transport costs for some building materials.

Building material costs are also affected by other domestic factors. Increased competition in other sectors of the economy, including the labour market, a reduction in taxes arising from government expenditure cuts and falls in real interest rates all have the potential to reduce building material costs indirectly during the next few years.

With respect to the building process, the residential building industry is dominated by small firms which build fewer than about ten houses each a year. They account for the bulk of housing construction. Moreover, the bigger firms employ sub-contractors to a large extent. The cost of entry into and exit from the industry is low. There is a continuing trend towards off-site manufacture in order to lower costs.

Builders' net margins are influenced by the economic cycle. High margins tend to arise when the economy and residential building activity grow strongly. Hence, in order to keep margins at an appropriate level on a sustainable basis, the government needs to avoid large swings in its economic policies.

The effect of the building code on direct building costs is significant. In the past, local authorities imposed excessive and very prescriptive requirements in approving buildings. Building permit fees were expensive in relation to the amount of work the local authorities did (up to 3 percent of the value of the permit), and there were often long delays before they could be obtained. After the Wahine storm in 1967, the building code became more stringent, even though much of the damage was apparently due to local authorities not enforcing the existing standards rather than to the standards themselves.

It should be noted that, while local authorities have been criticised heavily for the costs that the building code and its administration imposed, they were liable by law for any costs arising from inadequate or poorly administered building standards.

For a number of years, work has been under way to reform the building code. In 1990, the government approved the establishment of a nationwide performance-based building code which will be the responsibility of a national authority. It is intended to be flexible, admitting different solutions for meeting the required performance standards, rather than prescriptive. Builders will be able to use the private sector to certify that the code has been adhered to. The role of local authorities will be limited to checking the certificates.

However, there is considerable doubt as to whether the revised code will represent an improvement. The new code is still voluminous and very detailed. Its scope has been expanded beyond the primary areas of health and safety and certain third party effects, which may justify statutory regulation, to encompass issues such as amenity and energy efficiency which do not seem appropriate for a building code. Experience in England and Wales suggests that private sector certifiers will not emerge because of the number of trades involved, and that local authorities will continue to dominate the certification process.

The Building Industry Authority, which is responsible for the new national code (to be implemented following the enactment of the Building Bill), argues that it needs to ensure that minimum standards in building prevail. However, in an industry in which practical knowledge of each customers' requirements are important, and building performance guarantees are available (the Housing Corporation operates a building performance guarantee scheme and the Master Builders' Federation is intending to establish another), it is questionable whether such an extensive code, with its attendant direct and indirect costs, is required.

3.6 Labour Market

As mentioned, most small builders, who dominate residential housing construction, are self-employed and the larger firms often use contractors. It may, however, be the case that labour market regulation has inhibited the development of a different pattern of building firms which might be more efficient.

For building firms with significant numbers of employees, labour market regulation has been, and in some repects may remain, a handicap, particularly with respect to the employment of young and unskilled people. The award system has meant that firms have little incentive to invest in training, as they have limited scope to differentiate remuneration according to relative costs and productivity. Minimum wage regulation prevents employers hiring workers whose initial productivity is such that it is not profitable to hire them at the statutory rate of $245 per week. Ideally, firms would want to hire some unskilled staff at relatively low wages, to reflect their initially lower productivity and training costs, and pay them more subsequently when their productivity was higher. The result is that unskilled job seekers find it hard to secure jobs in the industry.

Another aspect of labour market regulation is the apprenticeship system. It is quite rigid. For example, the carpentry apprenticeship requires apprentices to become competent with respect to both residential housing and commercial and industrial construction. An apprentice who works only in the residential housing industry is therefore substantially overqualified. There may be a case for some form of official 'quality control' of qualifications for the purpose of validating claims about the level of competence and skills acquired. However, there seems no justification for prescribing by statute the contents or duration of a training programme or the terms and conditions of employment of those undertaking it.

Probably the most important impact of labour market regulation on the supply of housing has been felt indirectly. New Zealand's inflexible and monopolistic labour market arrangements have inflated costs throughout the economy, thereby raising the cost of a wide range of inputs into the industry. In addition, by prolonging the disinflation process, they have caused real interest rates to be much higher for longer than necessary.

Overseas experience in reducing inflation suggests that countries with flexible labour markets have been able to reduce inflation more quickly and with less unemployment and lower real interest rates than countries with labour market institutions like New Zealand's. Moreover, inefficient work and management practices, a lack of incentives to upgrade skills and uncompetitive pay rates have stifled the growth of employment and hence real incomes, and thus limited growth in the demand for housing.

The labour market reforms introduced in the Employment Contracts Act in May 1991 mark the realisation by the government that it could not ignore what has been one of the most severe economic policy problems in New Zealand. In addition, the Apprenticeship Act is being revised to make the apprenticeship system more flexible. However, it will still take some time to eradicate the legacy of past practices and attitudes. Moreover, the restrictions that have been maintained, in particular the mandatory personal grievance provisions, the retention of a specialist legal jurisdiction outside the civil courts and the high level of the statutory minimum wage relative to average earnings all mean that the labour market will not operate as well as it might.

With respect to occupational regulation, the previous government initiated a review of the regulation of electricians, gasfitters, engineers, architects, plumbers and drainlayers, and real estate agents. Some decisions were made before the 1990 election but were not implemented; they are currently subject to a review (this concerns engineers, architects and real estate agents). Reviews of the regulation of plumbers and drainlayers and lawyers have been discontinued. Only occupational deregulation with respect to electricians and gasfitters appears to be proceeding. It is disappointing that the new government appears less committed to reducing the costs imposed on housing by occupational regulation.

3.7 Rental Market

The stock of rental accommodation accounts for around 26 percent of the total stock of residential housing. About 22 percent of the rental stock is owned by the state through the Housing Corporation, and 60 percent by private landlords. The balance is provided by other government departments and local authorities or is rent-free accommodation, often supplied by employers to employees.

Underlying the government's policies of encouraging home ownership appears to be a belief that most New Zealanders have a strong preference for owning their own homes. Yet it is not immediately obvious why this should be the case. In Switzerland and Germany, for example, around 70 percent and 63 percent of occupants respectively rent their accommodation.

Renting has many intrinsic attractions. For a regular payment, renters may use accommodation without worrying about maintenance, rates, insurance and other homeownership problems. It is flexible, allowing renters to move town, to trade up or down, or simply to have a different lifestyle, relatively easily.

Renting may also have an advantage from a savings perspective. For most people, a house is their most important asset. However, it is also a relatively risky asset. Homeowners can easily lose all their savings by buying a house at the top of the market. Renting may allow people who are not wealthy to diversify their assets more effectively, so that they are better protected against the volatility in the price of any one asset.

Despite these intrinsic merits of renting, homeownership is the dominant form of accommodation in New Zealand. Reasons for this apparently include the freedom to modify one's home environment, independence and pride of ownership, security of tenure and financial factors.

The corollary of this is that the supply of rental accommodation in New Zealand is limited, it comprises mainly low cost housing and little is on the basis of long-term leases, except for Housing Corporation and local authority rental accommodation.

The private rental supply industry is fragmented. Few private landlords own many properties or rely predominantly on rent for their income; correspondingly, there is a marked absence of corporate involvement in supplying rental accommodation, except to employees.

Such a market structure is not in itself undesirable, as the split between rental and ownership tenure should be a matter of personal choice. However, the pattern in New Zealand may in part be a result of inappropriate government interventions. It would be a matter of concern if these significantly distorted consumer choice and the industry's structure, thereby reducing the productivity of housing from a national perspective. It is possible, for example, that the tax and other government-induced financial advantages of owner-occupation have been so strong that many people have felt unable to resist taking advantage of them. This is particularly so as politicians of all parties have made repeated commitments of support for homeownership, and have over the years fulfilled these commitments by directing credit towards mortgages, by seeking to keep nominal interest rates artificially low, and by providing extensive Housing Corporation subsidies. These actions have tended to imply that the price of houses would rise at a satisfactory rate over time, and generate an attractive rate of return to homeowners on an after-tax basis.

Thus, homeowners in New Zealand have been in a position not unlike that of farmers in the early 1980s; while they may perhaps realise that their investment is risky, and is dependent on continued political support, they would be reasonably confident that no government would remove that support. The decision by the previous government not to impose a capital gains tax on owner-occupied housing, in the event that the tax were introduced, is the most recent confirmation that such support is likely to be maintained.

In addition, the political risks of investing in residential rental property are considerable. Landlords are unpopular, and explicit rent controls have never been far from the agenda of some politicians, despite the disastrous effect they have had in countries like the United States in terms of increasing the number of homeless. Indeed, the below-market rents charged by the Housing Corporation may be seen as a form of implicit rent control, as private suppliers offering comparable types of accommodation are restricted from raising their rents by competition from the Corporation.

It is perhaps not surprising, therefore, that investment in residential rental property does not appear to have been an especially attractive investment over time. Furthermore, a National government was willing to introduce a discriminatory tax aimed at the capital gains made by landlords, which had the effect of significantly reducing the liquidity of newly-acquired properties. Repeal of this legislation was announced in the 1990 budget. The last Labour government introduced residential tenancy legislation which maintained and extended restrictions on the terms and conditions of contracts that may be entered into between landlords and tenants.

Most importantly, perhaps, the government has been willing, through the Housing Corporation, to increase the supply of rental accommodation at subsidised prices, apparently irrespective of the state of the rental market. For example, in 1989/90 the Housing Corporation acquired an additional 2,707 houses for rental accommodation, at rents approximately two-thirds the levels prevailing in the private sector, even though private sector rents did not appear to have been increasing generally.

This tends to suggest that, were government interventions neutral in their impact on the rental market, more New Zealanders would rent rather than own their accommodation and the private rental industry would expand and contain more businesses of larger size. These would perhaps lead to more long-term leases, more stable investment returns, a private sector more capable of responding to increased demand and hence more stable rents.

Hence, the rental market could be improved significantly by government interventions being reformed so as to remove the bias towards homeownership. This would involve not only making the tax treatment neutral, but would also mean that if the Housing Corporation were retained its programmes should be tenure-neutral. It would also be necessary to ensure that the Corporation's rental business was subject to similar financial disciplines to those faced by private sector landlords.

3.8 Housing Assistance

The discussion thus far has identified an extensive range of government policies which adversely affect the housing market, and reforms that may improve its efficiency. The next question to be addressed is whether there are valid efficiency and equity considerations, beyond those already discussed, which would justify direct housing assistance.

3.8.1 Economic Principles

One efficiency argument advanced for housing assistance relates to the inter-dependency of property values. When an individual improves his or her property, its value increases and this may simultaneously increase the value of nearby properties. Because an individual is unable to capture the latter benefits, it is sometimes argued that the investment in his or her property is likely to be less than a socially optimal amount. This spillover factor is common to many activities. It is questionable whether the benefits of government intervention to correct any real or perceived spillover effects in the housing market would outweigh the costs involved.

A more frequently advanced argument is that poor housing breeds crime, delinquency, disease and mental illness. In response to such arguments, Rosen has concluded that:

"It seems reasonable to believe that it is poverty associated with poor housing, rather than the housing per se, that causes these costly social problems".

General poverty is best addressed by broadly-based redistributive policies rather than by programmes aimed at addressing specific symptoms.

Housing assistance is sometimes advocated on the basis that people, particularly street kids and drifters, have irrational housing preferences. This view is also implied when it is argued that people choose poor quality housing because of their ignorance about its impact on physical and mental health and on educational achievement. In this latter case, the costs of gathering information on housing choices available may be a factor. These grounds for housing assistance also seem weak, given the importance of housing in overall household expenditure.

In the case of street kids and drifters, their alleged irrationality is not just a question of housing, but of their lifestyle as a whole. Whether society should attempt to modify it raises civil liberties issues. If society is committed to certain standards of personal liberty, it has to accept that some people who are not psychiatrically disturbed may choose lifestyles that others may regard as unusual or unsatisfactory, as long as this does not harm other people.

Another argument for housing assistance is that parents and guardians, acting as agents for their children, may choose to consume sub-optimal housing from the perspective of children, resulting in inferior educational attainment and inadequate standards of health. General income redistribution may not solve this problem, if parents or guardians appropriate this income to satisfy their needs. Housing assistance might be advocated in such situations as a means of reducing the costs of monitoring parents and guardians. However, the latter have incentives to act in the best interests of their children. The state provides for the rights of parents and guardians to be attenuated in serious situations such as when children are maltreated. Moreover, it is not clear that this problem is more compelling in housing than in other areas in which the state does not intervene so heavily, such as in early childhood care.

The Housing Corporation has expressed concern that the supply of housing may not be responsive to increases in demand due to production lags, which at times can be a year or more even where sections are developed. The concern is that slow adjustment results in rapid house price inflation to ration available houses to those prepared to pay, but does not call forth timely supply increases. Consequently, the Corporation has suggested that the government should provide houses at affordable prices to overcome apparent housing shortages. There are three common aspects to this line of argument.

The first concerns short-term homelessness, resulting from situations such as domestic violence and marital breakdowns. In these cases of urgent and immediate need for short-term shelter, there is a range of short-term accommodation available in the private sector, such as hotels, motels and boarding houses. The main cause of short-term homelessness appears to be the affordability of housing rather than the provision of the accommodation itself.

The second is the claim that private sector firms make mistakes in forecasting demand. However, the same can be said of the public sector. There is little evidence that the state is better at forecasting future housing demand than the private sector. The fact that the number of points required to obtain a Housing Corporation rental unit, and the average length of time which applicants spend on Housing Corporation waiting lists, varies from region to region tends to support this view. Moreover, state agencies not facing normal market disciplines may have incentives to forecast higher levels of demand than the private sector because the political embarrassment of under-forecasting and under-providing is large, whereas the costs of over-forecasting and over-providing may be hidden in the intricacies of government financing. Competing suppliers, on the other hand, have incentives to make optimal investments in their forecasting capabilities. If one firm under-forecasts, it will lose profitable opportunities to expand production, and may lose market share to those firms that make more accurate forecasts. Similarly, if one firm produces too many houses, profits are lost due to costs rising faster than revenue. Price movements and the backlog of orders provide information which firms use to plan their production schedules.

The third is the more general problem that new house construction may respond only slowly to situations in which actual demand outstrips forecast demand. However, as the major causes of delay in housing construction appear to be the time it takes developers to produce sections ready to be built on and the time required to acquire additional manpower and materials to construct houses, state provision of housing does not necessarily solve this problem efficiently. The optimal response for suppliers depends on the cost of carrying additional stocks of houses in relation to profits forgone from not meeting market demand. The cost of carrying surplus stocks of housing will include interest costs and the impact on housing prices if stocks have to be reduced, while the potential for lost profits will depend on, among other things, the responsiveness of house prices to supply shortages. The more responsive are house prices to shortages, the greater are potential forgone profits and the more worthwhile it is for house suppliers to add to surplus stocks. Similarly, the same incentives apply to property developers to produce efficient stocks of properties within the current planning and regulatory regime.

Moreover, increases in demand are not only met by new houses. They are also met by a more intensive use of the existing housing stock. Increases in demand cause house prices to rise, and hence to increase explicit and imputed rents in tandem, thereby giving people an incentive to use their housing more intensively. For example, houses will be turned into flats for renting; some people will postpone house purchases and share rented accommodation for longer; young people will defer moving away from home; some people will sell or let their holiday homes; some will sell their larger homes and move into smaller ones; relatives will move in; and so forth.

These alternative supply responses help to limit house price increases. The price mechanism coordinates the trade-off between maintaining surplus stocks of houses and more intensive use of the existing stock of houses. State provision of housing through purchasing from competitive suppliers does not avoid price rises any more than a gradual phase-in of cash assistance would. If the Housing Corporation were to approve a rush of orders for houses to be purchased, suppliers would face the same increase in demand as if the equivalent income transfers were made to Housing Corporation clients. Moreover, if the Housing Corporation directly produced houses, private suppliers would face reduced incentives to carry surplus stocks to meet housing shortages since the lower potential for higher prices would make it less worthwhile to do so.

Another point sometimes made is that the private sector is not interested in providing accommodation to low income groups. This argument seems difficult to sustain. The private rental market in New Zealand is geared towards servicing lower income people, because most higher income people tend to be owner-occupiers. The 1986 Census showed that the Housing Corporation was the landlord for just 62 percent of the unfurnished flats for which rents were between $30 and $40 per week. In all markets, some businesses specialise in targeting the needs of lower income groups. The question is fundamentally whether the returns are sufficient for the costs and risks involved in providing the accommodation.

Private sector landlords are sometimes criticised for discriminating against minority groups, such as Maoris, Polynesians and women. It is sometimes argued from this that direct housing assistance is required to offset discrimination against such groups. There is little doubt that some people do discriminate on racial, gender or other grounds. However, in a competitive housing market, discrimination on non-economic grounds against certain classes of people for mortgages and rental accommodation would have to be all-embracing to have a pervasive impact on the choices of the people involved. This is a crucial point which is often not appreciated. So long as there exist suppliers at the margin who are prepared to offer accommodation or mortgages without discriminating on non-economic grounds, housing costs cannot be affected significantly by any discrimination practised by other suppliers.

While general anti-discrimination laws have a role in controlling discrimination which is not based on valid economic considerations, effective competition among private suppliers provides the most potent protection for people at risk of discrimination. Competitive markets constrain discrimination on non-economic grounds by penalising those who practise it. This occurs because potentially profitable opportunities are not taken up by discriminating suppliers - instead those who discriminate according to race, religion or sex earn lower risk-adjusted returns on their property, at the margin and on average, compared to those who make good assessments of the risks they face.

On the other hand, 'discrimination' for economic reasons is likely to be widespread. In deciding to grant a mortgage, the lender takes a risk on the borrower's ability to repay the mortgage and on his or her trustworthiness in adhering to the contract entered into. In deciding to accept a tenant, the landlord takes a risk that the tenant will pay rent on time, will look after the property, will not offend neighbours or otherwise damage the landlord's reputation, will be reasonable and communicative in dealing with issues that need to be negotiated, and so on.

It is quite possible that people granting mortgages and letting accommodation will obtain information relevant to their risk assessment from the type of person that they are dealing with. For example, they may consider that young women are more likely than young men to have their income earning capacity disrupted to care for children. They may consider that foreigners with a poor grasp of English and with different cultural attitudes and habits would be more risky tenants. Reasonable subjective risk assessments of this nature are inevitable and normal, and occur in all aspects of economic activity. It seems inappropriate to regard them derogatorily as 'discrimination'.

Now that the market for mortgages is competitive, it is unlikely that discrimination on non-economic grounds is as widespread as is sometimes claimed. Moreover, discrimination on economic and non-economic grounds may be difficult to identify separately. If the private sector rental market were to become more competitive, in particular by the removal of actual and threatened forms of government intervention which affect it detrimentally, cases of discrimination would also be less frequent. A starting point in this regard would be a review of the Residential Tenancies Act.

Thus, it is difficult to provide strong arguments to support direct housing assistance. However, should the government wish to subsidise housing, one question that arises is whether this should be undertaken via income transfers, vouchers or in-kind assistance.

3.8.2 Income Transfers, Vouchers and In-Kind Assistance

Income (or cash) transfers allow complete freedom of choice to recipients to buy any goods and services they wish with the additional income. Vouchers typically restrict assistance to the purchase of specified goods or services. Concessionary mortgages granted by the Housing Corporation and accommodation benefits provided by the Department of Social Welfare could be classed as vouchers. In-kind transfers occur when the government provides goods or services itself. The Housing Corporation's rental housing programme is an example.

If information were costless and inadequate housing conditions were due to low income levels, cash, voucher and in-kind housing transfers would be equally efficient and effective at increasing housing standards and the welfare of recipients. An in-kind programme would be just as efficient and effective as a voucher scheme. This is because if information costs were zero, it would be costless to identify the features, cost and location of accommodation that would be chosen by recipients if vouchers were given instead, and it would be costless to monitor bureaucratic agencies to ensure that they acted in the interests of recipients.

Similarly, voucher assistance for housing would generate exactly the same pattern of housing consumption as if cash assistance were provided. This is because recipients of vouchers can spend their voucher income on housing and switch their existing income from housing to other goods. As there would be no restrictions on what recipients could spend their cash income on, recipients would choose only that amount of housing and other goods which they would buy if all income were provided as cash income. Consequently, both vouchers and cash transfers would increase recipients' welfare by equal amounts.

In these circumstances, only where the voucher would give free housing at levels that exceeded the quantum that recipients would choose if cash assistance were given would the vouchers not be equivalent to cash assistance. In this case, vouchers would reduce the recipients' welfare below that attainable with cash assistance because they would forgo consumption of other goods. It seems difficult to justify assistance policies which target specific goods at the expense of other goods which are equally valuable to recipients.

However, if monitoring costs are assumed to be positive and significant, the provision of vouchers or in-kind housing assistance may appear at first sight to be more effective in restricting housing assistance to the consumption of housing services. It is questionable whether this is in fact the case.

For example, one justification sometimes advanced for in-kind housing assistance is that taxpayers, who finance the programme, prefer their money to be spent on providing a certain standard of housing rather than, for example, on food, liquor, gambling and leisure. The idea that tying assistance will necessarily achieve that goal is an illusion, however. In-kind assistance does not ensure that taxpayers' money is spent on housing if recipients divert their income to buying other goods. The fact that welfare gains are less visible with income transfers should not be accepted as justification for in-kind assistance.

In any case, voucher or in-kind housing assistance may not ensure that recipients would consume the minimum standard of housing desired by the community. The co-existence of significant housing problems on the one hand with substantial welfare and housing programmes on the other, underlines this point. Moreover, in-kind assistance, to the extent that it restricts recipients from purchasing minimum quantities of other goods to satisfy goals that society is also concerned about - such as that of ensuring that children are well-clothed - defeats the intended purpose of in-kind restrictions.

It is sometimes argued that providing in-kind assistance helps to screen out people who do not really require or qualify for assistance by providing poorer quality goods and services than they would prefer. Moreover, by forcing the 'truly needy' to consume a certain basket of goods and services, consumption choices would tend to be reduced but the effectiveness of the programme may be increased, because expenditure may be better targeted. However, the monitoring costs of the Housing Corporation are likely to be lower with cash or voucher transfers, as private owners of rental units, for example, have better incentives to select appropriate tenants and to monitor the condition of their property and the behaviour of tenants. The optimal design of transfer programmes requires both consumption and programme efficiency to be taken into account.

The conclusion which we reach on the choice between cash, vouchers and in-kind transfers is essentially the same as that reached by Thurow:

"While it is not axiomatically true that cash transfers always dominate restricted transfers, the general economic case for cash transfers is strong enough that the burden of proof should always lie on those who advocate restricted transfers...With this in mind, I find, for example, the case for in-kind housing assistance unconvincing..."

More broadly, our assessment of the case for the provision of housing assistance is summarised by the following comments taken from Rosen:

"The main efficiency argument for subsidising housing is the existence of externalities [i.e. spillover effects]. However, the mechanisms through which these externalities work are not well understood and there is little evidence that they are quantitatively important. The redistributive rationalisation is equally weak. To the extent that society seeks to distribute income to the poor, the subsidies to owner-occupation are perverse because... they benefit mainly the middle- and upper-income classes. The in-kind subsidies involved in low-income public housing are inefficient in the sense that the poor could be made better off if the transfers were made directly in cash. Paternalism and political considerations seem to be the sources of this [housing assistance] policy."

3.9 Conclusion

Government policies affect housing in a wide variety of ways extending far beyond the details of explicit housing programmes. There is a great deal of scope for improving the housing market without broaching the question of direct assistance for housing. This can be done by improving overall economic management, the taxation system, the regulatory environment for planning and development, the variety of regulations which affect building costs (directly and indirectly), and government interventions in the private rental market. In general, the improvements envisaged would tend to lower the cost of supplying and purchasing housing, and enhance consumer choice. The case for direct housing assistance does not appear strong.

4. DIRECT HOUSING ASSISTANCE

4.1 Introduction

This section describes the main forms of housing assistance provided directly by the government. Information on recent trends in the amount of such assistance is also provided.

The Housing Corporation is the government's main agency providing housing assistance. Over 20 percent of all households are Housing Corporation mortgagors or tenants, or live in local authority accommodation, the erection of which was financed by the Housing Corporation.

The Iwi Transition Agency, which (in respect of the delivery of housing assistance) superseded the Department of Maori Affairs on 1 October 1989, provides housing assistance for people of Maori descent. Its programmes largely mirror those of the Housing Corporation.

The Housing Corporation's programmes and those of the Iwi Transition Agency fall into two main categories : general lending for housing purposes, and rental housing. These are examined separately below. The Housing Corporation provides insurance schemes related to its home lending operations which are also reviewed.

The Housing Corporation also lends to state servants on transfer, to armed forces personnel, to certain educational entities and for hotel development. These activities, which are undertaken on behalf of the government, are not primarily concerned with housing assistance. The state servants on transfer scheme, for example, is best categorised as a condition of employment. Such activities are not examined in this study.

Local authorities provide support for housing, and the Department of Social Welfare also provides housing assistance, largely through the accommodation benefit. These forms of assistance are also described below.

4.2 General Lending

4.2.1 Housing Corporation

The Housing Corporation's largest programme involves the provision of subsidised loans to individuals on low to modest incomes and to existing Corporation tenants for the purpose of acquiring their own homes.

This form of housing assistance dates from 1958. Initially loans were made to qualifying applicants at a non-reviewable interest rate of 3 percent per annum compared with a market rate of around 5.5 percent per annum. The 3 percent rate applied until 1974. Between 1974 and 1985, interest rates on new loans were gradually raised to reflect part of the large increase in private sector interest rates that occurred during the 1970s. Interest rates on existing loans became reviewable at increasingly closer intervals. Reviews have progressively been shortened from 5 years to 3 years and they are now undertaken annually or six monthly once the existing review date has been reached. The present interest rate regime (see below) was introduced in 1985 and 1986. It links the interest rate charged to the borrower's income and provides a market interest rate for borrowers who are deemed to no longer justify assistance.

The original home lending scheme was limited to the purchase of a new (i.e. not previously occupied) house by a first home buyer (i.e. a person who had not in the last five years held an equity interest in a house). The former condition reflected a concern to increase the stock of houses as a shortage of houses was seen to be a major problem. By 1974, the dual objectives of assisting low-income families to own a house and expanding the housing stock were in conflict as new houses were expensive compared with existing ones. Borrowers now have an unrestricted choice between buying a new or used home. In 1988/89 only 21 percent of new clients borrowed to buy or build a new house. The requirement that buyers be first home buyers was relaxed in the mid-1970s and abolished in 1979 in response to a rise in the incidence of family break-up.

In 1989/90, the Corporation authorised a total of around 26,200 housing loans amounting to about $643 million. This excludes the Corporation's agency business, such as advances to integrated schools, state servants on transfer, armed forces and Railways personnel, which is undertaken on behalf of the government.

Another indication of the significance of the Housing Corporation's lending programme is given by its residential mortgage advances. These primarily reflect the modest-income lending scheme but also include mortgages advanced under other schemes. At 30 June 1990 the Corporation's residential mortgage outstandings totalled $3.6 billion.

4.2.1.1 Modest-Income/Tenants Scheme

The following are the main features of the Corporation's current modest-income earners loan scheme:

(i) priority assistance is given to people occupying inadequate housing such as those in run-down, temporary, over-crowded or excessively costly housing; to those with a health problem; and to families whose income does not exceed the average wage;

(ii) the applicant must be a New Zealand citizen or a permanent resident of New Zealand;

(iii) assistance is available to people in need regardless of family size, dependants or previous homeownership. However, single people without dependants are not assisted unless they qualify for priority assistance;

(iv) there are no set loan limits. The Corporation satisfies itself that the borrower can afford the repayments. As a guideline a commitment of up to 30 percent of gross income is considered manageable;

(v) the borrower must contribute a minimum deposit of 12.5 percent of the house price in the form of cash or an equity interest in a section. If the borrower qualifies for the Homestart scheme (see below) the minimum deposit falls to 5 percent;

(vi) loans are made for a term of 30 years but the term is reviewable at any time after the first 10 years;

(vii) an application fee of 0.7 percent of the net loan is charged. This fee can be added to the loan amount and repaid over the life of the loan;

(viii) Housing Corporation rental tenants may borrow from the Corporation to acquire their rental unit (provided that it is not in demand) or another house. Income limits do not apply to this class of lending;

(ix) interest rates on new loans are constantly under review, but were reviewed annually prior to 1991;

(x) the prime or maximum interest rate on new loans as at April 1991 was 13.75 percent, reduced from 14.6 percent prevailing in February 1991. The government announced in December 1990 that in future the prime rate would be set equal to the mid-point of market rates rather than at the lower range of comparable market rates and that it would be automatically adjusted in line with the movement in private sector rates at later reviews. New clients have been required to pay the revised prime rate since 1 April 1991;

(xi) The Housing Corporation interest rate is reduced where the household's gross income during the last 12 months is under $431 a week ($22,412 a year). The following table shows the interest rate payable by household income and the extent of the interest rate subsidy compared with the weighted average rate charged by the nine major private sector providers of housing finance in March 1991 of 13.7 percent.

Modest-Income Lending

Income-Related Interest Rates

At 31 March 1991

Subsidy

Gross Average Interest Rate Percentage

Weekly Income Percent Per Annum Point

Up to $250 7 6.7

$251 to $330 9 4.7

$331 to $380 11 2.7

$381 to $430 13 0.7

Source: Housing Corporation of New Zealand.

The Corporation has discretion to modify these conditions if the circumstances of the applicant justify special assistance. A borrower may seek temporary assistance should he or she face a loss of income. In such cases, the Corporation can suspend principal or interest payments or extend the term of the loan for a maximum of six months when a review would be undertaken.

The Housing Corporation reported in February 1988 that 94 percent of its applicants were single income families of whom 43 percent were social welfare beneficiaries. Half of the applicants were couples with dependants and 40 percent were solo parents. The proportion of Housing Corporation mortgagors who were beneficiaries after three years of homeownership was under 10 percent and the proportion applying for relief from market interest rates was 20 percent.

In February 1988, 31 percent of Housing Corporation clients were paying the then maximum interest rate (17 percent per annum), up from 1 percent two years earlier. As at 31 December 1987, two-thirds of the value of outstanding loans was subject to an annual interest review and 46 percent of the amount outstanding was subject to interest rates of between 1 and 10 percent per annum. The average interest rate charged at 31 March 1989 on new loans made in 1988/89 was 11.17 percent compared with the then prime interest rate of 15.5 percent. On an average loan amount of $42,500, this suggests an average tax-free subsidy of around $1,840 per borrower in the first year.

An indication of the size of the Corporation's modest-income scheme and recent trends in the programme are given in the following table:

Modest-Income Lending

Authorisations

1984/85 to 1989/90

Year   Number Amount Average

Percent Percent Loan (1) Percent

Change $m Change $ Change

1984/85 6,013 - 177.4 - 29,500 -

1985/86 8,629 44 303.7 71 35,200 19

1986/87 9,670 12 354.1 17 36,600 4

1987/88 9,560 -1 377.6 7 39,500 8

1988/89 9,423 -1 400.2 6 42,500 8

1989/90 6,189 -34 267.8 -33 43,300 2

(1) Approximate only

Source: Annual Reports of the Housing Corporation of New Zealand, 1984/85 to 1989/90.

The above table shows the impact of the Labour government's expanded housing programme which was introduced in November 1984. The number of loan approvals grew by 61 percent between 1984/85 and 1986/87 before stabilising. In 1989/90, the Corporation's lending under this scheme was substantially cut back without significant changes to the scheme's eligibility rules. Even so, the Corporation reported that only 74 percent of the number of households assisted met the Minister of Housing's criteria of being in serious housing need. In the previous year only 50 percent of households assisted were deemed to have satisfied that criterion.

Lending to tenants has grown strongly since the moratorium on the sale of Corporation rental units (which had been imposed in August 1984) was lifted in October 1986. The Housing Corporation attributed the reduction in the number of loans authorised in 1989/90 to a fall-off in the number of Corporation tenants able to move out of rental accommodation. The scheme also helps tenants to acquire houses other than Corporation rental units.

Tenants Lending Scheme

Authorisations

1984/85 to 1989/90

Number Amount Average Loan (1)

Percent Percent Percent

Year   Change $m Change $ Change

1984/85 1,594 - 54.0 - 33,900 -

1985/86 688 -57 23.7 -56 34,450 2

1986/87 407 -41 15.3 -35 37,600 9

1987/88 555 36 24.4 59 43,960 17

1988/89 1,913 245 97.5 299 50,970 16

1989/90 1,194 -37 55.4 -43 46,400 -9

(1) Approximate only

Source: Annual Reports of the Housing Corporation of New Zealand, 1984/85 to 1989/90.

4.2.1.2 Homestart

Homestart is a scheme whereby prospective first home buyers are assisted to accumulate the deposit necessary to acquire a mortgage from either the Housing Corporation or another lender. This scheme was introduced in 1986 in place of the Home Ownership Savings Scheme, the building industry suspensory loan scheme and the Family Benefit capitalisation scheme, which previously provided substantial assistance to first homeowners.

The main features of the Homestart scheme up to December 1990 were as follows:

(i) families, couples and single people (over the age of 26) who were on low to middle incomes and who wished to buy or build a first home could qualify;

(ii) qualifying applicants received a loan for five years on which interest of 3 percent a year was charged but capitalised. Provided that the house was not sold, the loan amount plus capitalised interest was repayable at the end of five years. The loan could be extended at a higher interest rate;

(iii) the maximum value of the house which could be purchased was set by area. A price limit of $120,000 applied in the districts referred to below;

(iv) a Homestart advance could be taken up with or without a Housing Corporation mortgage;

(v) applicants purchasing a new house could borrow a supplement to help meet GST;

(vi) the amount of loan assistance available from 1 October 1989 reflected the income level of the applicant and the price of houses. The following table applied to higher priced districts such as Auckland, Tauranga, Taupo, Wellington, Lower Hutt and Nelson.

Homestart

Income-Related Loan Amounts

Higher-Priced Housing Districts

Sole Applicant's Family Deposit GST

Income Income Assistance Assistance

Up to 23,000 Up to 32,000 12,000 1,000

23,000 - 24,000 32,000 - 33,000 10,700 890

24,000 - 25,000 33,000 - 34,000 9,400 780

25,000 - 26,000 34,000 - 35,000 8,100 670

26,000 - 27,000 35,000 - 36,000 6,800 560

27,000 - 28,000 36,000 - 37,000 5,500 450

28,000 - 29,000 37,000 - 38,000 4,200 340

29,000 - 30,000 38,000 - 39,000 2,900 230

30,000 - 31,000 39,000 - 40,000 1,600 120

Over 31,000 Over 40,000 - -

Source: Housing Corporation of New Zealand.

This scheme has grown significantly since its introduction despite a reduction in authorisations in 1989/90, as the following table shows.

Homestart Scheme

Authorisations

1986/87 to 1989/90

Average

Number Percent Amount Percent Loan(1) Percent

Year   Change $m Change $ Change

1986/87 7,651 - 68 - 8,900 -

1987/88 13,194 72 120 77 9,100 2

1988/89 16,381 24 162 35 9,900 9

1989/90 12,837 -22 124 -24 9,700 -2

(1) Approximate only

Source: Annual Reports of the Housing Corporation of New Zealand, 1986/87 to 1989/90.

In 1985/86, authorisations for the schemes which were replaced by Homestart amounted to around $55 million.

In November 1990, the Minister of Housing expressed a concern that up to 30,000 homeowners would be unable to repay their Homestart loans when the loans first become repayable. In response to this problem, and to better target the scheme, the government announced that from 19 December 1990 the following conditions would apply to new Homestart loans:

(i) income-related interest of between 7 percent per annum and the prime rate would be payable on the loans. The income-related interest rate scale set out above for modest-income lending applies;

(ii) progressive repayments of principal would be required. These would be calculated on the same basis and term as the Corporation's modest lending scheme;

(iii) after five years, all outstanding interest and principal would be repayable.

In addition, the budget for Homestart was cut by $30 million and $50 million in 1990/91 and 1991/92 respectively. Most of the reduction in 1990/91 reflected a 19 percent fall in authorisations in the first half of that year.

4.2.1.3 Second Chance/Refinancing

In the recent past, the Corporation's programmes have been primarily aimed at assisting people to acquire their first home. The second chance/refinancing scheme, which was introduced in 1974, provides finance for people who would not qualify under the modest-income/tenants scheme or Homestart. It is available in situations where a family is in danger of a forced sale, where a family is suffering severe hardship as a result of high interest rates or where a marriage has broken down. The Corporation stated that it "has assumed the role of lender of last resort" under this scheme. The Corporation's prime interest rate is charged on second chance/refinance loans but a rebate of 15 percent is applicable where there is evidence of hardship.

The significance of the second chance/refinancing scheme is evident from the following data:

Second Chance/Refinancing

Authorisations

1984/85 to 1989/90

Average

Number Percent Amount Percent Loan(1) Percent

Year   Change $m Change $ Change

1984/85 761 - 16 - 13,000 -

1985/86 1,924 152 46 187 23,700 182

1986/87 1,552 -19 37 -19 23,800 -

1987/88 1,614 4 42 13 25,900 8

1988/89 2,323 43 73 73 31,600 22

1989/90 3,574 53 121 65 33,900 73

(1) Approximate only

Source: Annual Reports of the Housing Corporation of New Zealand, 1984/85 to 1989/90.

This scheme has expanded rapidly over the past two years as funding has been transferred from the modest-income and tenants schemes.

4.2.1.4 Other Lending Schemes

The Housing Corporation also provides the following smaller schemes:

(i) home improvement loans are available to undertake essential repairs (for example, after a flood) to expand necessary space or to provide accommodation for dependants. Up to $20,000 may generally be borrowed, although larger loans are possible, but the total amount borrowed against the property cannot exceed 90 percent of its value. The Corporation's prime interest rate is payable unless total housing costs are more than 30 percent of gross income;

(ii) elderly homeowners of limited means (those whose only income is the Guaranteed Retirement Income and whose cash assets do not exceed $16,000 for a single person (or $19,000 for a couple)) can borrow up to $20,000 to undertake home improvements. Interest and principal repayments may be capitalised during the life of the homeowner;

(iii) sweat equity loans are available to applicants with building skills (or who have a friend or relative with building skills) who intend to renovate an older house or build a house from a kit set. The Corporation either offers houses or buys a house selected by the applicant and rents it to the applicant on a low rent for six months while it is being repaired. The Corporation pays for materials. At the end of the period, the Corporation sells the house to the applicant at a price which takes account of the buyer's input;

(iv) under a home-swap scheme, the Corporation will buy large houses from elderly persons and in turn sell them a smaller, low-maintenance house. A loan may be provided if there is a net cost to the client. A similar scheme - Better Use Loans - provides help for elderly persons who cannot rehouse themselves. Disabled and elderly people may also qualify for a loan to acquire a relocatable cottage. This scheme was introduced in 1989/90;

(v) loans are also available for cooperative housing, women's housing projects, innovative housing projects and other special projects. These cover proposals which do not fall within other schemes but which assist in housing people on low or modest incomes;

(vi) under the Papakainga scheme, the Corporation makes loans available for the construction of housing on multiple-owned Maori land;

(vii) a new scheme called New Horizons was introduced in 1989/90 with the objective of encouraging Auckland households who could not afford homeownership to move to a lower cost area. Under this scheme homeowners can establish their eligibility for a Corporation loan and for a Homestart loan prior to making a move to a new area.

The size of the above programmes in 1989/90 is set out below:

Miscellaneous Loan Schemes

Authorisations

1989/90

Amount (1)

Scheme Number $m

Home improvement 1,370 12

Equity sharing/sweat equity 39 2

Tenancy saving scheme 32 -

Tied accommodation 11 -

Housing for the elderly/disabled 66 11

Better use/relocatable cottages 178 3

Special lending 192 39

Papakainga 246 11

New horizons 124 5

Building industry suspensory 175 1

_____ ____

Total 2,433 84

_____ ____

(1) Approximate only

Source: Annual Report of the Housing Corporation of New Zealand, 1989/90.

4.2.2 Iwi Transition Agency

The Iwi Transition Agency provides comparable modest-income, Homestart and refinance/second chance lending schemes to those provided by the Housing Corporation. In addition, the Agency administers the following special policies:

(i) Wharetapiri, a programme under which accommodation is added to an existing home by way of attached or detached units to prevent the break-up of large or extended families;

(ii) the Whareawhina policy, which provides for dwellings to be built on or adjacent to a marae in an environment where whanau and hapu support are available. The dwellings are likely to be marae-owned;

(iii) Maatua Whangai, which is a joint programme involving the Agency, Justice and Social Welfare. It aims to prevent at-risk Maori youth from being placed in institutions. Families participating in the scheme are eligible for housing assistance;

(iv) Papakainga housing, which provides for housing to be erected on multiple-owned land.

Spending on lending schemes administered by both the Housing Corporation and the Iwi Transition Agency is essentially determined by demand. While priority is given to applicants with serious housing need, other applicants who qualify might be required to wait for funds but they will not be refused assistance. If spending were to become excessive, the government would be required to alter eligibility criteria in order to reduce demand or to provide additional funds.

4.3 Insurance Schemes

The Housing Corporation provides on its own account two insurance schemes, in addition to offering insurance against fire and other risks in respect of homes and their contents on an agency basis for a private insurer. The schemes relate to mortgage repayment insurance and protection against faulty materials and poor performance by builders. These schemes are reviewed below.

4.3.1 Mortgage Guarantee

The Housing Corporation has offered a mortgage guarantee scheme since 1953. The scheme indemnifies approved home lenders against loss caused by defaulting mortgagors. As the scheme has been offered free of charge in the past, the Housing Corporation has had a virtual monopoly of the housing mortgage guarantee market since its inception.

In July 1990, the Housing Corporation scheme was refocused to target mortgage guarantees to households on modest incomes seeking homeownership. The Corporation aimed to achieve this by only offering to guarantee mortgages on those homes which had a value lower than the average value of houses sold in each of twelve regional districts. In addition, an application fee of $100 was introduced.

The guarantee is available in respect of the borrower's primary place of residence. The main criteria for the homeowner to gain approval are that he or she:

• meets the lender's requirements for loan finance

• is not under age (i.e. a minor)

• is not currently bankrupt

• is not subject to summary instalment orders

• obtains an independent valuation of the property

• uses the mortgage finance to buy, build or improve the mortgagor's home, ownership type flat or boarding house (with no restriction on the age, location or construction of the house)

• has replacement fire and earthquake insurance on the property (with some exceptions for boarding houses).

The loan must also be secured by a mortgage registered under the Land Transfer Act 1952 and the lender must not charge interest rates which exceed the Housing Corporation's prime rate for residential lending by more than two percentage points on first mortgages and four percentage points for second and subsequent mortgages. The maximum guarantee available varies according to the twelve regional price limits. The gross loan must not exceed 90 percent of the current market valuation of the property or the regional price limit, whichever is the lower.

Following a mortgagee sale, the Housing Corporation will pay the principal outstanding, provided that this does not exceed the original amount guaranteed, plus up to one year's accrued interest and costs.

The Housing Corporation stated in its 1989/90 annual report that the contingent liability in respect of its Mortgage Guarantee and Buildguard schemes could not be readily determined but that an actuarial assessment indicated that the liability could be approximately $30 million.

4.3.2 Building Performance Guarantee

Under the building performance guarantee (Buildguard) scheme, the Corporation provides protection against faulty materials and poor performance by builders. Homeowners pay a fee based on the value of the house with a maximum of $286 and receive cover for up to $200,000.

In the December 1990 economic statement, the government announced that Buildguard and the Mortgage Guarantee schemes would be placed on a full cost recovery basis from 1 July 1991.

4.4 Rental Housing

4.4.1 Housing Corporation

The second main area of direct housing assistance relates to the Housing Corporation's rental housing programmes.

The extensive construction of houses for state rental purposes commenced in 1936. The account formed for the purpose, the Housing Account, was financed by low interest loans (until 1956, generally at around 1 percent per annum). In 1949 emphasis was placed on homeownership and tenants were encouraged to buy their houses on favourable loan terms. By March 1965, over 65,000 units had been built or purchased by the Corporation, and nearly 20,000 units had been sold.

Since 1973, the housing needs of applicants for Housing Corporation rental units have been assessed on a points system. The Corporation makes the initial assessment of each applicant's position. Housing allocation committees, which comprise independent people drawn from the local community, generally finalise the assessment. Available properties are allocated to those applicants in the relevant area who are deemed to have the most urgent need.

As from 2 April 1990 a new pointing system has been applied to applicants. It provides for a maximum of ninety points to be allocated under six categories. A summary of the system is given in the table below.

The checklist for inadequate accommodation (see table 1c) includes the number, age and sex of adults and children that are housed, the number of bedrooms, the number of bedrooms assessed to be required and whether other facilities (e.g. kitchen) are shared. For each additional bedroom that the household requires, four points are awarded. If facilities are involuntarily shared an additional four points are awarded. The checklist also includes an assessment of the physical condition of the applicant's existing accommodation.

Income is the most heavily weighted of the six categories. It directly accounts for fifty of the ninety possible points. In contrast to the Corporation's loan scheme, income limits applicable to its tenants are calculated on an adjusted after-tax basis. Total net weekly household income is calculated as follows:

Income after-tax or net benefit

Plus average overtime

Family Support

Guaranteed Minimum Income

Less child care costs

Total net household income.

Rental Housing

Pointing System

Maximum Maximum

Points Points

1. Existing accommodation 20

a) Not a dwelling 20

b) Emergency accommodation 18

c) Inadequate accommodation 20

(see checklist)

d) Insecurity of tenure 2

(for use when c is below

maximum)

e) Location 2

(for use when b or c is

below maximum)

2. Income (ability-to-pay) 30

Maximum points are awarded where

income is below or equal to the

appropriate benefit level (s) for

household composition. Decrease by

1 point for every $10 extra net

income over benefit level.

3. Rent (affordability) 20

Threshold equals the appropriate

Corporation rent for income (from

2 above). Points accrue when actual

rent exceeds this amount.

4. Time with priority 5

The full five points accrue for

applicants whose applications

have been confirmed but who are

waiting to be housed six months

on from attaining that status.

5. Ill-health 5

Categories range from slight to

serious and urgent.

6. Discretionary points 10

To be awarded by full housing

allocation committee, which will

specify reasons given.

Maximum Total Points 90

Source: Housing Corporation of New Zealand (1990b).

The total net household income (calculated above) is compared with the total base income level (including Family Support) set by the Corporation. The latter varies according to the number of adults and dependants included in the household (see table). The total base income for a single adult with two children, for example, is $293 a week plus Family Support. If the total net household income is less than or equal to the total base income level the maximum of thirty points applies. If total net household income exceeds total base income, then the number of points awarded is thirty less one point for every $10 by which total net household income exceeds total base income.

The total base income level has been set with respect to the level of social welfare benefits. Thus where the household is receiving a single benefit and is earning no additional income, the maximum of thirty points applies. If there are more than two adult applicants or partners, the difference between the net unemployment benefit rate for a couple and that for a single person is added for each extra adult.

Base Weekly Income in $ (1) (2)

Dependants Adults on tenancy agreement plus partners

1 2 3 4 5 6 7 8

0 162 270 357 444 531 618 705 792

1 255 312 399 486 573 660 747 834

2 293 334 421 508 595 682 769 856

3 315 356 443 530 617 704 791 878

4 337 378 465 552 639 726 813 900

5 359 400 487 574 661 748 835 922

6 381 422 509 596 683 770 857 944

7 403 444 531 618 705 792 879 966

8 425 466 553 640 727 814 901 988

(1) If income is solely Guaranteed Retirement Income then the current income amounts available under that scheme are used.

(2) Add Family Support where applicable.

Source: Housing Corporation of New Zealand.

The approach as described includes a revised treatment of Family Support. Family Support was previously included in household income and excluded from total base income. However, as some applicants were found not to be obtaining Family Support despite their entitlement, Family Support has been added to both total net household income and total base income. The effect of this change is to enhance low-income earners' chances of receiving a Housing Corporation unit and to reduce the effective marginal tax rate applicable to households containing such persons who wish to apply for a unit.

The Corporation's income assessment is intended to measure the applicant's ability to pay. Its assessment of rent payable measures the affordability of the applicant's existing accommodation. The calculation is similar to that described above. First, the level of rent which the Corporation would charge on the applicant's total household income (as calculated above) is computed. Next the actual rent paid by the applicant is reduced by the Department of Social Welfare Accommodation Benefit (where applicable). The Housing Corporation's assessed rent and the adjusted rent paid are each divided by the net household income. For every 1.7 percentage points that adjusted actual rents paid (as a percentage of household income) exceed the assessed rent (as a percentage of household income) one point is awarded.

Discretionary points are awarded in exceptional circumstances which are not able to be adequately measured within the first five factors. They could, for example, be awarded where there is extreme stress caused by the use of shared facilities despite a lack of overcrowding or where households are split between more than one house.

Applicants who receive fifty or more points are deemed to have an urgent housing need. The Corporation would search for a suitable unit for such applicants. Other applicants could be expected to wait a longer time before gaining a unit.

As noted, Housing Corporation rents are set according to the after-tax income of the applicant, subject to that rent not exceeding the assessed market rent for the relevant unit. The following illustrative rents applied from 1 April 1991.

Housing Corporation Rents

Household Class Household Income Weekly Rent

Net of Tax

$ $

Single adult (25 years and over)

sickness beneficiary with no children 135.22 34.00

Domestic purposes beneficiary 227.93 57.00

with one child

Unemployed married couple 293.88 73.00

with two children

Married couple on Guaranteed 288.10 72.00

Retirement Income

Single person on Guaranteed 187.26 47.00

Retirement Income

Wage earner 372.00 118.00

Wage earner (1) (existing tenant) 450.00 167.00

(1) A unit in Auckland having a high value.

Source: Housing Corporation of New Zealand.

The above rents are generally lower than those which applied from 2 April 1990. The benefit reductions implemented from 1 April 1991 have been accompanied by a reduction in income-related rents.

Housing Corporation tenants are required to pay a bond equivalent to one week's rent at the market rental for the property and one week's rent in advance. Social welfare beneficiaries and other persons can apply to the Department of Social Welfare for a special needs grant to help meet these costs. This assistance depends on the applicant's income and assets. There is also provision to rebate rents where the tenant is over-committed.

Rents are reviewed every twelve months. Maximum rent increases are calculated according to net household income excluding Family Support. The maximum increase in rent payable by social welfare beneficiaries is $10 a week. There is no limit on the amount of the rent increase for other households. In December 1990, the government announced that, from 1 April 1991, tenants paying market rents would have their rents aligned with the mid-point of comparable market rents and not the lower range of market rents as had been the practice.

Limited tenancies were introduced on 1 October 1976. They provided for a six yearly review of tenancies. This provision was, however, terminated on 31 July 1984. Currently, there is no time or income limit on tenancies. If a tenant's circumstances improve, rents are increased to a market level. Tenants paying market rents are encouraged to purchase their own house. After six years, they can apply to buy their unit which could be sold depending on whether the Corporation desires to retain or sell the particular unit. Housing Corporation loans are available to tenants to acquire Housing Corporation rental units or other houses.

In addition to an income test, Housing Corporation applicants are subject to an asset test. All assets except household furniture and appliances and one car are included. The present guidelines allow a maximum of $25,000 for the Auckland (excluding Whangarei), Manukau, Henderson, Porirua and Lower Hutt (excluding Masterton) districts and $20,000 elsewhere. Applicants whose assets exceed the limit can still have their eligibility determined by a housing allocation committee.

A current or recent interest in a property does not automatically disqualify an applicant provided that the asset test is met. Special conditions may apply, for example, in the case of a marriage breakdown. The Corporation may also decline to provide a rental unit to former tenants or mortgagors who are in default.

A Housing Corporation study of new rental clients in 1987/88 showed the following family types:

New Rental Clients

Family Type

1987/88

Average

Income

Family Type Percent $/week (1)

Childless couples 20.3 295

Solo parents 51.2 250

Person without

dependants 9.8 180

Couple and children 16.8 275

Other 1.9 280

_____

Total 100.0

_____

(1) Rounded

Source: Housing Corporation of New Zealand (1988) pp. 80-86.

The income source of new rental clients in 1987/88 was as follows:

New Rental Clients

Source of Income

1987/88

Source of Income Percent

Social welfare benefits 74.7

Wages 24.8

Other 0.5

____

Total 100.0

____

Source: Housing Corporation of New Zealand (1988) pp. 80-86.

In order to house tenants, the Housing Corporation has built up a large stock of rental units. At 30 June 1990 it amounted to 67,753 units. At the 1986 Census it accounted for around 22 percent of the total rental housing stock and almost 6 percent of the total housing stock in New Zealand. The government valuation of the rental stock at March 1989 plus subsequent additions to 30 June 1990 at current market value, amount to around $3.55 billion. The trend in the Housing Corporation's rental stock is set out in the following table:

Rental Housing Stock

1964/65 to 1989/90

Stock at Units Purchased

Year End of Year or Constructed

1964/65 47,504 4,400

1969/70 51,798 5,323

1974/75 52,388 3,803

1979/80 60,368 7,364

1984/85 57,547 5,395

- - - - - -- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 1985/86 59,145 5,662

1986/87 60,600 7,065

1987/88 62,094 7,939

1988/89 64,521 9,545

1989/90 67,753 2,707

Source: Housing Corporation of New Zealand (1988) and Annual Reports of the Housing Corporation of New Zealand, 1987/88 to 1989/90.

The Housing Corporation's stock of rental units peaked in 1981 at 60,465 before tenants were given enhanced opportunities to acquire their units. This was part of a policy designed to better target the existing stock of units and was associated with moves to increase the rents of existing tenants who were no longer in need of assistance. In 1984, the incoming Labour government placed a moratorium on the sale of units to tenants. This moratorium was lifted in October 1986 but each unit sold to a tenant is to be replaced and the sale of a unit is only approved if the unit is not in demand. These decisions, together with an expansion in the state house acquisition programme, resulted in a turnaround in the rental housing stock which is now at a record level. In its December 1990 economic statement, the government announced that expenditure on the acquisition of rental units would be reduced by $40 million, halving the planned acquisition from 2,400 to 1,200 units. In addition, the government announced its intention to maintain the rental stock at around 70,000 units.

4.4.2 Iwi Transition Agency

Since 1965 over 300 flats have been established under the Kaumatua flats programme to house elderly people near, or adjacent to, marae. The maximum rent charged is 25 percent of net weekly income.

4.4.3 Local Authority Housing

According to the 1986 Census, local authorities own 16,524 rental dwellings of which 15,627 were let on an unfurnished basis. Wellington and Auckland City Councils are probably the second and third largest owners of rental accommodation in the country. The average rent charged for local authority unfurnished flats was $32.09 a week compared with $27.06 for Housing Corporation units and $86.44 for privately owned flats. The average rent charged for furnished flats owned by local authorities was $36.53 a week compared with $86.90 for privately owned flats. (Housing Corporation units are all unfurnished.)

Local authority housing tends to fall into two categories. The first is community housing which is available to anyone, although preference is generally given to applicants deemed to be in need. The Household Expenditure and Income Survey 1988/89 suggests that around one third of local authority housing is let to employees. The second category is pensioner housing, generally comprising blocks of small flats, the construction of which is assisted by the provision of subsidised Housing Corporation loans. In 1989/90, for example, 65 loans totalling $10.9 million were approved for local authority and community groups to acquire pensioner flats.

Following substantial rises in rates, some local authorities have reviewed their rental housing operations. The Auckland City Council, for example, is increasing some rents to market levels and it is examining the possibility of selling part or all of its housing stock. As a consequence of the rent increases, a number of properties have become vacant.

4.5 Department of Social Welfare - Accommodation Benefit

The Department of Social Welfare may supplement beneficiaries' and non-beneficiaries' income by providing an Accommodation Benefit. The benefit is targeted at people with limited income and cash assets who have high accommodation costs. Housing Corporation clients, other than beneficiaries, are generally ineligible for this benefit although persons who have Homestart or home improvement loans may qualify.

To qualify for the benefit a sole person's weekly rent must exceed $41 while that of a married couple or single person with children must exceed $68 a week. The relevant rates for boarding costs are $61 (single) and $102 (married/parents). Beneficiaries who own their own home may qualify if their outgoings exceed $49 (single) or $82 (married/parents) a week.

The amount of the benefit payable is 50 cents for each dollar that the applicant's accommodation costs exceed the benchmark costs (referred to above) except for those who pay board. The maximum level of the benefit is $41 (single) or $68 (married/parents) a week. People with a disability or older people in residential care may receive an additional $20 a week.

A single person may have cash assets of up to $2,700 and a married couple/parent up to $5,400 before the amount of the benefit is abated. Each $100 of cash assets in excess of these amounts reduces the benefit by $1 a week. Applicants with cash assets in excess of $8,100 (single) or $16,200 (married/parent) are ineligible for the benefit.

The Accommodation Benefit is also income tested. For beneficiaries, the first $80 a week of income over and above the benefit reduces the Accommodation Benefit at the rate of 25 cents in the dollar. Income earned beyond this level does not affect the amount of the Accommodation Benefit provided that the person remains on a social welfare benefit. For low-income earners, the Accommodation Benefit is reduced by 25 cents for every $1 earned in excess of $154.47 (before-tax) a week for 15 to 17 year olds, $192.32 for a single adult, $342.57 for a married person with or without children, $262.99 for a sole parent with one child and $284.79 for a sole parent with more than one child.

Low-income earners in the above income ranges would generally be paying a marginal income tax rate of 28 percent. The combined effect of income tax and the phase-out of the Accommodation Benefit would be to produce an effective marginal income tax rate of 53 percent (assuming no other income-related assistance such as Housing Corporation rents or concessional interest rates). Persons in receipt of Guaranteed Retirement Income or a veteran's pension face a reduction in the benefit at the rate of 25 cents in the dollar for the first $80 a week of income other than their Guaranteed Retirement Income or pension. The benefit is not further abated but is withdrawn if their income exceeds the following limits:

$299 a week - single adult

$454 a week - married with no children

$458 a week - married with children

$376 a week - sole parent with 1 child

$398 a week - sole parent with 2 or more children.

At 30 June 1990, 106,431 people were receiving an Accommodation Benefit. The total cost of the benefit was $2.3 million a week ($2.1 million in March 1989) which is equivalent to $1,133 a year for each recipient. A breakdown of recipients is given below.

Accommodation Benefit

At 30 June 1990

Weekly

Class of Value

Primary Benefit Number $ percent

Guaranteed

Retirement Income 6,935 146,430 6

Widows 927 21,039 1

Invalids 8,785 281,110 12

Domestic purposes 25,987 744,083 32

Sickness 7,726 168,803 7

Unemployment 52,737 908,896 39

Training 2,911 35,683 1

Non-beneficiaries 423 12,533 1

_______ _________ _____

Total 106,431 2,318,577 100

_______ _________ _____

Source: Department of Social Welfare.

5. EVALUATION OF DIRECT HOUSING ASSISTANCE

5.1 Introduction

The purpose of this section is to evaluate the main forms of direct housing assistance provided by the government, which were described in section 4, in terms of standard efficiency and equity criteria.

5.2 Objectives

In order to evaluate the government's housing programmes it is necessary to examine their objectives. In its final submission to the Royal Commission on Social Policy, the Housing Corporation wrote:

"Access to housing of an adequate standard is essential if the standards of a fair society, as defined in the Royal Commission's Terms of Reference, are to be met".

It went on to state that:

"While the majority of people are able to meet their housing needs in the market without assistance, a significant proportion are not. If these people are to be housed, and if the standards of a fair society are to be realised, then the Government must continue to play a major role in housing. In addition to this equity consideration, the Government will want to be involved in housing because housing is a merit good. That is, society as a whole benefits if all citizens have access to adequate housing".

In the Corporation's view, the main factors constraining access to housing were "affordability, the existence of discrimination and inadequate supply". Supply problems were caused by an absolute shortage in the stock of housing and a lack of sufficient housing of "the right type and tenure in the right place for particular groups". This problem was attributed to the fact that "the market does not respond quickly to economic, social and demographic changes".

The former Minister of Housing's 1989/90 policy statement stated that his priority was "to work on practical policies to relieve serious housing need" arising from overcrowding, substandard housing, temporary accommodation, unaffordable housing and social deprivation. While that statement was a desirable attempt to specify more precisely the aims of the government's housing policy, the assessment of housing needs is still largely subjective. Substandard housing, for example, depends on the specification of minimum acceptable facilities and the assessment of priority for rental assistance requires points awarded under different headings to be added to arrive at a meaningful total score which is used to rank applicants. The aggregation involved in this process has been questioned.

Leaving aside the question of the difficulty of assessing whether the Corporation has achieved the objectives set by the government, the Corporation itself noted that in 1988/89 and 1989/90 only 50 percent and 74 percent respectively of new borrowers under the modest-income lending scheme met the Minister's requirement of being in serious housing need. This is one of the Corporation's largest programmes.

The Housing Corporation's conclusion that the majority of people are able to house themselves is consistent with the government's general approach to welfare assistance. This is based on the view that most people can provide for themselves but that the state should provide safety net assistance for those who are unable to do so.

A number of questions raised by the Corporation's objectives were examined earlier in this study. These included the grounds for housing assistance, such as discrimination and the adequacy of the supply of housing services, as well as arguments for and against vouchers and in-kind transfers. In our view, none of these arguments provides a compelling case for direct housing assistance in the longer term. This view is also supported by an analysis presented below of the main programmes in terms of targeting, the level of assistance provided and the incentive effects of such assistance.

5.3 General Lending

5.3.1 Targeting

The targeting of general lending programmes has been improved since 1984. The main improvements involved the establishment in 1986 of a prime interest rate which broadly equates with the market interest rate on first mortgage lending to prime clients, and the annual or six-monthly adjustment of interest rates on existing loans to the prime rate for households earning average or better incomes. Initially, a borrower who satisfied the Housing Corporation's lending criteria received a subsidised interest rate regardless of his/her subsequent income. Moreover, the amount of the subsidy fluctuated widely as private sector interest rates were more volatile than those charged by the Corporation.

Another recent improvement was the phase out of the Home Ownership Savings Scheme and the Building Industry Suspensory Loan Scheme. The former scheme could provide the equivalent of a 20 - 30 percent rate of return on savings for many people who were relatively well-off first home borrowers. The latter provided additional assistance for those buying a new house.

In the December 1991 economic statement, the government moved to realign the prime interest rate with the mid-point of private sector rates, to reduce the attractiveness of the Homestart scheme and to put the Buildguard and Mortgage Guarantee schemes on a more commercial basis. These changes can also be expected to improve the targeting of the Corporation's lending programmes.

Despite these measures, the Housing Corporation's general lending schemes continue to provide significant assistance to households that are moderately well off. The following observations support this conclusion:

(i) as at 31 March 1991, the prime rate was not charged until a household income of $431.00 a week was reached. The Household Expenditure and Income Survey 1988/89 suggests that almost 35 percent of all households earn incomes equal to or less than $431.00 a week;

(ii) there is some evidence in the Housing Corporation's analysis that households with temporarily low incomes qualify for housing assistance, which is then phased out gradually. This follows from the fact that 40 percent of Housing Corporation applicants under the modest-income earners scheme were found to be single income beneficiaries but that the number of borrowers who were beneficiaries after three years of homeownership was less than 10 percent. Secondly, only 20 percent of borrowers applied for relief from the prime interest rate, suggesting that most borrowers were ineligible for a subsidised interest rate after three years. These results probably reflect the high incidence of unemployment among young workers and the rise in income as young workers gain experience. The latter might include applicants who are undergoing training (e.g. apprenticeship, tertiary study) but who subsequently earn at least average incomes;

(iii) Corporation tenants who are paying market rents or are approaching market rents are assisted to buy their unit or another house. While this has the benefit of freeing up Corporation units for reallocation to higher priority applicants, it has the effect of providing additional subsidies to households that are at least moderately well off;

(iv) the existing accommodation of applicants for Housing Corporation loans is not assessed to determine whether they are in serious need in terms of the criteria set for housing assistance. The Corporation reported that in 1989/90 some 30 percent of new low- to modest-income borrowers did not satisfy the criteria laid down by the Minister of Housing.

5.3.2 Level of Assistance

Under the general lending programmes, assistance is provided by way of an implicit subsidy comprising the difference between market interest rates which would apply to the borrower during the term of the loan and those which are charged by the Housing Corporation. An indication of the aggregate amount of assistance provided in 1988/89 (March year) and 1989/90 (June year) is given in the following table.

Estimated Average Mortgage Interest Subsidy

1988/89 and 1989/90

1988/89 1989/90

(March Year) (June Year )

Mortgage interest charged

by the Housing Corporation $000 (1) 388,185 390,317

Mortgages outstanding at

the start of the year $000 3,262,610 3,570,482

Mortgages outstanding at

the end of the year $000 3,532,632 3,740,499

Average amount of mortgages

outstanding $000 (1) 3,397,621 3,655,490

Average interest rate earned

by the Housing Corporation (%) 11.4 10.7

Estimated average market

interest rate (%) (2) 16.0 15.7

Estimated amount of subsidy $000 156,290 182,775

(1) Includes non-residential mortgages. These account for around 3 percent of the average amount of mortgages outstanding.

(2) Monthly weighted average of the prime interest rate charged by the main providers of housing finance as reported by the Reserve Bank of New Zealand.

This calculation under-estimates the subsidy to the extent that Housing Corporation borrowers would not be considered to be prime borrowers by private sector lenders and to the extent that the interest rate on existing private sector loans is charged at above their prime rate for new lending. The subsidy would also be increased (or decreased) to the extent that the Housing Corporation's origination fee is less (more) than that charged by private sector lenders.

On the basis outlined, the mortgage interest subsidy is estimated to have increased from around $156 million in 1988/89 (March year) to almost $183 million in 1989/90 (June year). This arose because the average interest rate charged by the Corporation appears to have fallen faster than that charged by private sector lenders, possibly reflecting a reduction in the average income of the Corporation's borrowers. An average interest rate of 10.7 percent in 1989/90, as calculated above, suggests that the average household income of Corporation borrowers was between $331 and $380 a week.

The amount of assistance provided to each borrower varies in the following ways:

(i) the larger the loan amount borrowed or outstanding, the higher the subsidy;

(ii) the higher private sector interest rates are relative to the Corporation's rate, the greater the subsidy. In the recent past, the subsidy has changed significantly because private sector interest rates have risen or fallen without a corresponding adjustment to Corporation rates;

(iii) the lower the borrower's gross household income (i.e. before-tax), the higher the subsidy per dollar borrowed. Household income includes wages and overtime but excludes fringe benefits.

The relationship between interest rates and income increases the subsidy payable to low-income people compared with those on higher incomes. However, households on low incomes borrow lower amounts than higher income families because of their inability to service the loan. Thus, it is not clear that the largest subsidies are paid to households on the lowest incomes.

5.3.3 Incentive Effects

The Housing Corporation's general lending programme has important effects on the incentive to work and on tenure choice. These aspects of the programmes are discussed below.

5.3.3.1 Labour Market

The level of assistance under general lending programmes and Homestart is income- related. As a consequence, these schemes can be expected to adversely affect the willingness of potential applicants and existing borrowers to work. They can be expected to encourage households to minimise income prior to lodging an application for a loan, as the level of assistance provided decreases as income rises. This might encourage potential applicants to defer promotion, to refuse to work overtime and to register for a social welfare benefit (e.g. an unemployment benefit). Moreover, in the case of households containing two or more adults, one or two members may cease working on a full- or part-time basis. Existing Housing Corporation borrowers, who are subject to an interest rate below the prime rate, are also discouraged from working, as their interest rate increases when their income increases.

The combined effect of marginal rates of income tax, the phase-out of the Family Support tax credit and income-related interest rates act as a considerable disincentive to work. Between a household income of $13,000 p.a. and $22,412 p.a. the Housing Corporation interest rate rises from 7 percent to the current (March 1991) maximum of 13.75 percent. Over this income range, the low-income earner rebate is phased out, producing a marginal income tax rate of 28 percent. In addition, the Family Support tax credit for the first child is phased out on family incomes (exclusive of the credit) in excess of $17,500 and up to $28,580. The relevant income ranges for more than one child are slightly higher.

An example illustrates these factors. Consider a single income household with one child earning $360 a week and with $40,000 owing on a Housing Corporation mortgage. This family would be paying 11 percent on their mortgage ($88 a week in principal and interest) and would face a marginal income tax rate of 46 percent (comprising 24 percent on income, 4 percent for the phase-out of the low-income earner rebate and 18 percent arising from the phase-out of Family Support). If the household's income increases to $381 (an increase of 5.8 percent) a week, its interest rate would rise to 13 percent and the impact on disposable income would be as follows:

$ p.a.

Increase in gross income 1,092

Tax on incremental income 502

Net of tax income 590

Additional interest and principal payments 737

Net reduction in disposable income after mortgage

payments in year one 147

The effective marginal tax rate in this example is 114 percent. It is a dramatic example because the household is assumed to earn close to the maximum income allowed for a 11 percent interest rate prior to the increase in income. The household could subsequently increase its income to $430 a week without facing an additional interest cost. The effective marginal tax rate rises at the income level at which a higher interest rate applies. More extreme examples are possible. However, this example does illustrate the disincentive to earning income which applies around the income points at which higher interest rates come into effect.

There is an incentive for prospective and existing borrowers to seek remuneration in a form which does not affect the recorded level of household income as far as the Housing Corporation is concerned (e.g. fringe benefits). Prospective recipients of Housing Corporation assistance may also gain by supplying false information to the Corporation and by withholding relevant information about their circumstances.

The Corporation's general lending programmes previously applied to first home seekers only and as such were a significant impediment to labour mobility. This provision has now been removed. Nevertheless, the loan amount available to applicants is mainly limited by the applicant's ability to service the loan. Because housing is more expensive in the larger centres, the overall quality of housing attainable with assistance differs from centre to centre. The regional variations in loan limits and in Homestart assistance do not fully reflect differences in housing costs. Consequently, the general lending programme tends to favour homeowners in smaller centres and can be expected to have some adverse impact on labour mobility.

5.3.3.2 Tenure Choice

There has been a strong bias in favour of homeownership over private rental accommodation from the commencement of the government's general lending programmes. As discussed earlier, this has retarded the development of the private sector rental market and reduced consumer choice.

5.4 Mortgage Guarantee Scheme

5.4.1 Targeting

The Housing Corporation announced in January 1990 that it would focus its Mortgage Guarantee scheme on the lower half of the residential property market. This had the potential to achieve significantly tighter targeting of the scheme to low and medium income families. Initially, this was to be achieved by setting regional limits on guarantees based on the median house sale price for each of the twelve regional districts specified by the Housing Corporation. Such a policy would have left 50 percent of the housing market available for the private sector to offer to guarantee against default.

Instead, before the new policy started on July 23 1990, the Housing Corporation substituted the average house sale price for the median house sale price in setting its maximum level of guarantee. This switch in formula leaves only 10-15 percent of mortgages on residential properties available for the private sector to guarantee. Moreover, the applicant's housing situation, income and assets are not taken into account in deciding whether to provide a guarantee. Therefore, targeting of the Housing Corporation scheme has in practice been loose and has not focused on assisting people in serious housing need. Indeed, the rationale for providing a subsidised guarantee scheme for households which do not qualify for other assistance because they are on moderate to high incomes is unclear.

The decision to place the Mortgage Guarantee Scheme on a full cost recovery basis from 1 July 1991 may help to correct this problem. However, a further review of the scheme currently being undertaken by the Corporation with its approved lenders may signal a delay. Much will also depend on the allocation of overheads in determining what is the full cost of the scheme. To be fully comparable with private mortgage insurers, these should include reinsurance premiums, taxation, the servicing of capital and an allowance to the Crown for the implied/expressed guarantee provided.

5.4.2 Level of Assistance

The Housing Corporation charges a fixed application fee of $100 whereas private sector agencies would be likely to charge fees of up to $1,000. On this basis, the implicit subsidy to approved lenders ranges up to $900 per applicant depending on the amount of the mortgage and the risk of default by the borrower.

5.4.3 Incentive Effects

Mortgage guarantees insure lenders against default by mortgagors. The effect of such guarantees is to shift the risk of default to the insurer who generally then has an incentive to manage that risk, including managing the problems of moral hazard and adverse selection. For customers seeking mortgage finance, particularly for those where the ratio of the loan to the value of the asset exceeds 80 percent, mortgage guarantee schemes may enable them to obtain finance at cheaper interest rates, since the lender faces less risk, but at the cost of paying at least a portion of the application fee passed on by the lender.

A borrower is likely to take up a mortgage guarantee if its cost is less than the additional interest cost payable if a guarantee is not taken. In practice, guarantees prove most useful where they enable borrowers to take up larger loans than otherwise acceptable to lenders.

The Housing Corporation Mortgage Guarantee scheme has two main effects on incentives. First, the Housing Corporation has little incentive to manage risk and to establish appropriate risk categories as its survival does not depend on profit-making activities. Since the Corporation is taking on the risk of guaranteeing loans, lenders have reduced incentives to ensure that the risk is effectively assessed and that accurate applications are made. Furthermore, should arrears arise, the lender has a reduced incentive to address the problem promptly. Uninsured lenders, on the other hand, have a larger incentive to reject risky applications. Lenders in this situation may decide to offer the loan at higher interest rates (and take on the risk themselves) or to decline to lend at all. However, the Housing Corporation scheme runs the risk of becoming a lax means of providing loans to people who are at risk of defaulting. This could be especially serious if borrowers realise they can subvert the risk assessment process by submitting inaccurate applications.

Secondly, the $100 application fee encourages risk taking in excess of the true cost of that risk. Some borrowers who at the margin may have been deterred if they had to pay the real cost of the guarantee scheme may not be deterred by the $100 charge. Since high risk loans are often those which have a high debt level in relation to security, individuals who are most likely to default on loans are not encouraged to reduce their risk level by, for instance, accumulating larger deposits. Rather than assisting low-income earners, the Housing Corporation guarantee scheme may encourage high-risk/low-income groups into debt burdens beyond their means to a greater extent than would otherwise be the case. Thus, the scheme may reduce the incentive for those with less secure sources of income to choose accommodation which incurs least risk of financial loss.

5.5 Rental Housing

5.5.1 Targeting

The Corporation's rental programme has also become better targeted in recent years as a tenant's ability to pay is now reviewed subsequent to the initial allocation of a unit. Moreover, excess demand for rental units has helped to ensure that applicants most in need are allocated units. Housing Corporation tenants are generally less well off than many of the Corporation's low-to modest-income borrowers.

The Treasury noted in 1984 that the then current housing stock was probably sufficient to meet the most urgent housing needs. One problem has been to free up existing units so that they can be reallocated to higher priority uses or to charge full market rents where households no longer justify assistance. While the move toward market rents has helped, there is little doubt that some households on the waiting list would be assessed to have a greater priority than some existing tenants who are still in receipt of subsidised rents. There is also a wide disparity in the number of points required and time taken to gain a unit in different districts.

5.5.2 Level of Assistance

Under current policies, there will always be excess demand for Housing Corporation rental units because tenants receive substantial assistance by gaining a unit. The following analysis based on the Corporation's annual accounts suggests that the Corporation earns a low rate of return on its units:

Return on Housing Corporation Rental Units

1987/88 to 1989/90

1987/88 1988/89 1989/90

$m $m $m

Rental income 171 212 233

Costs and expenses

other than interest 142 172 186

and depreciation __ __ __

Operating surplus 29 40 47

__ __ __

Government valuation of

properties 2,761 3,247 3,548

Operating surplus:

value of properties (percent) 1.1 1.2 1.3

The approximate level of subsidy in 1989/90 (June year) can be estimated as follows:

Estimated Subsidy on Rental Units

1989/90

Assumed Market Required Actual Estimated Avg Subsidy

Yield(1) Surplus Surplus Subsidy Per Household(2)

Percent $m $m $m $

6 213 47 166 2,500

8 284 47 237 3,570

10 355 47 308 4,640

12 426 47 379 5,710

(1) This is the pre-interest and tax capitalisation rate based on the latest government valuation of the properties (including, in respect of 1989/90, the Housing Corporation's estimate of the market value of additions since the latest (1989) valuation). This overstates the return actually earned as government valuations are updated on a rolling five yearly basis and thus the actual market value of the units could be expected to exceed their government valuation.

(2) Approximate only. An average of 66,400 households are assumed to have benefited from the subsidy. This number has been derived by taking the stock of units at 30 June 1990 and reducing it to reflect units which were vacant for part of the year.

A commercial investor in rental housing would require a total return (rental income plus change in the market value of the properties) which reflected the risk-free rate of interest plus a premium for the risk inherent in investing in rental accommodation. The Corporation's pre-interest and pre-tax income return is substantially below the commercial norm. A pre-tax income yield of 10 percent would put the subsidy for 1989/90 at around $308 million a year (or $4,640 per household) compared with $285 million in 1988/89.

The best measure of the subsidy would be to compare Housing Corporation rents with market rents for equivalent properties. Unfortunately this information is not available. The following Auckland examples illustrate the extent of the subsidy which could be involved in the main cities. The data relate to the position in April 1991. The market rent is the rent which the Housing Corporation valuers estimate the Corporation's units would command in the market.

Selected Tenants and Rent Levels

Auckland City

Housing Description

Corporation Net Income of Net Rent Market

Tenant of Tenant Accommodation $p.w. Rent(1)

Single adult 135 1 bedsitter in 41 110

sickness beneficiary central city

block

Married Guaranteed 288 1 bedroom flat 72 125

Retirement Income in central city

block

Domestic pur- 228 2 bedroom 57 170

poses flat in

beneficiary with Mt Roskill

1 child

Married un- 294 3 bedroom 73 230

employed couple house in

with 2 children Mt Roskill

Wage earning 450 3 bedroom 166 240

household with house in

2 children Mt Albert

(1) Market rent as assessed by Housing Corporation valuers.

Source: Housing Corporation of New Zealand.

The market rent of Housing Corporation units depends on their location. Thus within Auckland City, the market rent on a two bedroom flat could rise from $170 a week in Mt Roskill to $200 (Auckland Central) and to $220 (Eastern Suburbs). The market rent of a three bedroom house in the Eastern Suburbs could be $280 a week.

The estimated subsidies are substantial. In the first case listed, the rental subsidy is equivalent to an increase in pre-tax income of more than 50 percent from $8,259 to $12,979 p.a. Clearly, the above examples would not apply in smaller centres, nor are they necessarily typical of those applying in the main centres.

5.5.3 Incentive Effects

5.5.3.1 Labour Market

As with the Corporation's general lending programme, its rental housing programme can be expected to have an adverse impact on the incentive to work and earn a market income. With a beneficiary's net-of-tax income high in relation to the incomes of lower paid and average earners, the subsidy available through rental accommodation can make it unattractive to take paid employment. Furthermore, a rise in market incomes results in an increase in rents, thereby discouraging tenants who are in paid employment from seeking additional income.

The rent formula for an initial Corporation rental unit applicant is as follows:

(i) 25 percent of net income up to the net rate received by a married couple in receipt of ;

(ii) 35 percent of additional net income up to the level of the average all persons weekly wage;

(iii) 65 percent of additional net income. This last step would imply an effective marginal tax rate of 74 percent for recipients subject to the low-income earner's rebate and assuming no Family Support tax credit. Where this credit applies, the effective marginal tax rate would rise to 92 percent.

The rent formula for an existing tenant who is a beneficiary is subject to a maximum annual increase of $10 a week. There is no limit on the increase that is applicable to other tenants except that income-related rents cannot exceed the assessed market rent for the particular unit.

Because of the demand for existing rental units, tenants are discouraged from moving to another locality unless they are still deemed to be in urgent need of housing assistance or there is a surplus of state rental accommodation in the desired locality. Thus the Housing Corporation rental programme is likely to impede labour mobility.

5.5.3.2 Rental Market

The adverse implications of the Housing Corporation's programme for the private rental market have already been discussed.

5.6 General Considerations

The foregoing discussion examined some of the economic effects of the Corporation's main programmes separately. The following more general points are also important:

(i) the level of assistance provided to Housing Corporation borrowers and tenants varies greatly depending on their personal circumstances. There is little reason to believe that the level of assistance relates closely to the need of particular clients or that the level of assistance provided in each form is equal. Moreover, the complex rules which apply, particularly in the rental area, make it difficult to determine accurately the efficiency and equity effects of current policies;

(ii) as is generally the case with assistance provided in-kind, little regard is had to consumer preferences. This is particularly so in respect of rental assistance. The high level of subsidy biases many peoples' preferences. On the other hand, a significant number of the poorest households, in terms of income, do not benefit from government housing assistance. Presumably, they prefer to obtain their own accommodation rather than to seek a cheaper Corporation unit, or they may be on the Corporation's waiting list;

(iii) to a large extent, housing assistance is an important adjunct to social welfare assistance. Beneficiaries are best placed to gain access to a rental unit and, where they can support a loan, they receive the most favourable terms. In effect, housing assistance provides a significant supplement to social welfare support. This point is well illustrated by the recent reduction in some benefit levels which lead to a reduction in income-related rents and interest rates for some Corporation clients;

(iv) the level of assistance provided by the Accommodation Benefit appears to be substantially less than the average subsidy provided to Housing Corporation tenants;

(v) because living costs are highest in the main cities and social welfare benefit levels are uniform throughout the country, the resulting income pressure is reflected in the demand for Housing Corporation rental accommodation (and the accommodation and special needs benefits). It may well be better to recognise variations in the cost of living and to address this in the benefit system than to provide additional housing assistance;

(vi) regional differences in housing costs are inadequately recognised in existing housing programmes. There is a case for adjusting, to a greater extent than currently, the loan amount available to reflect the costs of housing in particular areas;

(vii) because the determination of housing needs is essentially subjective, the construction of housing units in particular localities and the allocation of units to households is open to lobbying;

(viii) once the government decides to provide substantial assistance for an activity, it is faced with a difficult trade-off between high effective marginal tax rates faced by the recipients as the assistance is phased out or a high fiscal cost (and hence higher effective marginal tax rates on all households) of a more gradual phase-out or of universal assistance. In general, a high effective marginal tax rate may have tolerable efficiency costs if it applies to a relatively narrow income band which most people can expect to pass through and if it does not apply to a large proportion of the labour force. The difficulty raised by Housing Corporation interest and rental income formulae is that they apply over wide income bands which cover a large proportion of the labour force (including workers who are likely to be relatively easily discouraged). The programme can therefore be expected to have a significant adverse impact on willingness to work. This is particularly so in relation to social welfare beneficiaries. The only practical way to address this problem is to lower the level of assistance provided.

5.7 Directions For Reform

As discussed earlier, it is our view that voucher and in-kind housing assistance should be phased down and that income transfers, perhaps designed to better reflect differential housing costs, should generally replace direct government housing assistance. This would give consumers greater choice between spending on housing services and other goods and services, thereby helping them to maximise their welfare. It would also provide greater choice of housing services and be neutral between homeownership and rental accommodation. Under these conditions direct housing assistance would not be required as a general rule.

So long as specific housing subsidies are provided, we believe that existing assistance should be reformed with the following objectives:

(i) to further improve the targeting of assistance so that only those in serious need are subsidised;

(ii) to reduce the adverse disincentives of existing assistance.

Desirable reforms (within the current policy and institutional framework, and beyond those announced in December 1990) would include the following:

(i) terminate the Homestart scheme thereby requiring prospective homeowners to provide a larger deposit before qualifying for a loan;

(ii) tighten the eligibility rules for general lending so that households with temporarily low incomes and middle income families with no serious housing need do not qualify;

(iii) reduce and make consistent the level of assistance available under both rental accommodation and general lending programmes, thereby reducing the disincentive impact of these schemes;

(iv) give greater weight to housing costs in particular areas in setting levels of housing assistance;

(v) cap the rental housing stock at 60,000 units and introduce policies aimed at achieving a better use of the existing stock and at a better distribution of units among regions. In particular, limited tenancies should be reintroduced;

(vi) review existing Housing Corporation clients' eligibility more frequently with a view to better targeting assistance;

(vii) remove the current preferential treatment of Housing Corporation tenants for Housing Corporation loans. Such tenants would only be able to borrow if they meet the criteria applicable to other applicants;

(viii) abolish the Mortgage Guarantee and Buildguard schemes.

The government's housing policy changes, announced in December 1990, move in the appropriate direction in our view, but should be seen as a first step only.

6. THE ROLE OF THE HOUSING CORPORATION

6.1 Introduction

The business operations of the government have come under increasing scrutiny during the 1980s and early 1990s and reforms aimed at improving their performance have been implemented. In this section, the role of the Housing Corporation is examined in the light of the principles against which state businesses have generally been evaluated.

While it might be argued that the Housing Corporation is not a business, because of its social role, it is nonetheless difficult to conclude that an organisation with such a large lending and renting operation does not constitute a business in some sense. The previous government's decision to reconstitute the Housing Corporation as a department does not alter this conclusion.

6.2 Examination of State Businesses

Examinations of the recent performance of state businesses have led to the following broad conclusions:

(i) their objectives tend to be unclear and are often in conflict;

(ii) competitive advantages conferred on them artificially (e.g. by regulation or policy) impair efficiency;

(iii) they often have excessive operating costs and earn low commercial returns;

(iv) they make poor quality investment decisions. This, as well as (iii), flows from the relatively weak incentives and disciplines facing their managers and from inappropriate political interventions in management decisions.

The main steps required to enhance the performance of state businesses and improve their efficiency are listed below:

(i) the reform of regulations which affect the industries in which they operate to remove policy-induced barriers to competition;

(ii) the removal of artificial competitive advantages enjoyed, and competitive impediments borne, by them;

(iii) a clarification of their objectives by separating their commercial and regulatory roles, with the latter being undertaken by departments;

(iv) the establishment of performance targets for commercial operations to improve accountability;

(v) a transfer to private ownership to enhance the incentives facing their managers.

6.3 The Housing Corporation's Role

The Housing Corporation is a major state business. It is a large financial intermediary and property manager. In addition, it provides policy advice on housing matters, undertakes various regulatory tasks, particularly in relation to tenancies, and, in assessing housing needs, acts as a welfare agency. One unusual feature of the Corporation's operations is that it provides rental housing assistance on an agency basis. The Housing Account, through which this activity is undertaken, has not yet been included in the Corporation's own accounts nor has it been consolidated in the accounts of any department.

It would not be surprising if the problems identified with state businesses in general also applied to the Corporation. It has several objectives, some of which are potentially in conflict. It is, for example, charged with tendering advice on housing policy generally, yet it is the major market participant in both the rental and mortgage markets. The Corporation also has artificial advantages over its competitors which include the following:

(i) its services are heavily subsidised by the government;

(ii) it administers legislation which regulates the rental market in which it is the largest single participant;

(iii) it is not required to earn a return on its assets. Its operating costs appear to have grown strongly in recent years, thereby raising the question of whether its cost efficiency has declined;

(iv) its debt has been guaranteed by the government. This was changed in 1990, and the Corporation is now funded exclusively through the Treasury.

As noted elsewhere, the Corporation's involvement in the rental and mortgage markets, and in providing mortgage guarantees, can be expected to have affected the quality and quantity of services which would have been provided by the private sector, thereby reducing consumption choices available to New Zealanders.

6.4 Directions for Reform

If the government's assistance to housing were provided indirectly via general income maintenance programmes, as advocated in this report, rather than by voucher or in-kind assistance there might be no need to retain the Corporation as a Crown agency. Its regulatory and advisory roles could be transferred to another government agency, such as the Ministry of Commerce. Its existing mortgage portfolio could be sold or allowed to run off, and the rental stock offerred for sale. An alternative would be to privatise the Corporation as an ongoing business or a series of businesses. Any residual welfare activities, such as emergency shelter, might be passed to local and community based welfare agencies or the Department of Social Welfare.

If, for whatever reason, the government wished to continue to provide direct housing assistance, there would be merit in reforming the Corporation. The main aim of this reform would be to enhance efficiency in the mortgage and rental markets and to provide better incentives to raise the Corporation's performance in participating in these markets.

The new business would be charged with meeting appropriate commercial objectives and be required to compete on neutral grounds with other entities. The approval of applications for assistance, whether by way of subsidies on rent or mortgage interest, should, of course, not be the responsibility of the rental or lending business. Applicants should be permitted to utilise their subsidy by taking up a Corporation rental unit or a private sector unit. Similarly, the interest subsidy could be applied to a Corporation or private sector mortgage.

This approach would require assistance levels to be explicit so that private sector entities were reasonably certain of the support provided by the government. Assistance should also be provided in such a way that it could be used in either purchasing or renting a house (i.e. it should be tenure neutral). This would enable the delivery of in-kind assistance to be separated from the ownership of rental units and the provision of mortgages and thus facilitate competition in both markets.

The sale of the Corporation's businesses, or its assets, should enable government debt to be reduced by perhaps $6 billion. This would reduce the risk premium embedded in interest rates and improve the government's fiscal position. Whether or not it would be worthwhile selling the businesses rather than the assets would need further analysis. The government would need to be satisfied that there was value in the existing businesses in a competitive context if it decided to sell its interests by this means.

7. CONCLUSIONS AND RECOMMENDATIONS

There has been a tendency in recent years to view housing policy largely in terms of subsidised Housing Corporation mortgages and rental accommodation. This study has examined housing policy in a broader context. It is evident that there is a wide range of measures that the government and its agencies can and should adopt to improve housing in New Zealand. Implementing these measures on a comprehensive basis would lead to:

• an improvement in consumer choice of housing

• a reduction in the cost of housing

• an improvement in the quality of housing

• a reduction in welfare dependency and unemployment

• a reduction in net government expenditure and indebtedness

• an improvement in the productivity of housing from a national economic perspective and hence a higher level of national output and income.

The principal measures that need to be adopted are as follows:

Economic Management

• the 0 - 2 percent inflation target by 1993 needs to be achieved and sustained

• the government needs to eliminate the fiscal deficit and subsequently to achieve fiscal surpluses, in order to establish a more stable macroeconomic climate.

Finance

• the government should allow any subsidies related to housing finance to be delivered by any financial institution, not solely the Housing Corporation.

Taxation

• as part of an ongoing strategy of reducing government spending relative to GDP and of moving towards lower tax rates on a broad tax base, the government should, over time, review the income tax treatment of owner-occupied housing

• the government should consider extending GST to owner-occupied and rental housing services

• the government should treat owner-occupied and other housing on a comparable basis to other classes of investment with respect to any taxation of capital gains

• the local authority rating system, the funding of the fire service and the Earthquake and War Damage scheme should be reformed with the aim of reducing their adverse impact on the costs of housing.

Planning and Development

• the government should undertake a further phase of resource management law reform to encourage greater use of market solutions to planning and development issues, and to make the quasi-judicial process less cumbersome

• local authorities should adopt least cost regulatory and market solutions to planning and development issues within the prevailing legislative framework

• local authorities should act to minimise the delays they cause in the planning and development process by virtue of their administrative and regulatory role.

Building Costs

• the government should advance the programme of import liberalisation and other deregulation to increase competition in areas affecting building costs

• the Building Industry Authority should re-examine the minimum building standards embodied in the building code.

Labour Market

• the government should continue to review legislation and regulation with the objective of making labour market reform thoroughgoing and comprehensive. This should include a review of any unnecessarily prescriptive features of the Employment Contracts Act and the role of the Employment Tribunal and Court

• the government should give fresh direction and impetus to the current review of occupational regulation with the objective of achieving appropriate liberalisation.

Rental Market

• the government should announce its intention to allow the development of a more broadly-based and sustainable market for rental accommodation by ensuring that subsidised state rental housing does not crowd out private sector provision

• the Residential Tenancies Act should be reviewed

• local authorities should examine whether the provision of rental housing is an appropriate function for them.

Assistance to Housing

• the government and local authorities should discontinue provision of direct or in-kind assistance to housing

• the government should consider whether general income maintenance programmes should be modified to a greater extent than at present to take account of differential housing costs.

Institutional Reforms

• the government should sell either the loans of the Housing Corporation or its lending businesses

• the government should sell the Housing Corporation rental units

• the regulatory and advisory activities of the Housing Corporation should be transferred to another government agency, and residual welfare activities relating to housing should be transferred to community welfare agencies or the Department of Social Welfare

• if specific housing assistance programmes are maintained, the policy reforms suggested in section 5.7 should be considered.

APPENDIX I

An Estimate of the Cost of Selected Forms of Housing Assistance

In this Appendix an estimate of the cost of the main components of government assistance to housing for 1988/89 and 1989/90 is set out.

Cost of Selected Housing Assistance Programmes

1988/89 and 1989/90

1988/89 1989/90

$m $m

Accommodation Benefit(1) 108 121

Subsidy on Housing Corporation mortgages(2) 156 183

Subsidy on Housing Corporation rental units(3) 285 308

Tax forgone on imputed rents(4) 512 452

_____ _____

Total 1,061 1,064

_____ _____

(1) See section 4.5. This estimate assumes that the weekly payments at the end of March 1989 and June 1990 are indicative of the average payments made during 1989/89 and 1989/90 respectively.

(2) See section 5.3.2 for the derivation of these estimates.

(3) See section 5.5.2.

(4) See Appendix 2.

The above calculation omits the cost of the Housing Corporation insurance schemes and some subsidies involved in the sale of Housing Corporation rental units. Similarly, it omits the cost of subsidies provided by the Iwi Transition Agency and local authorities.

APPENDIX II

An Estimate of the Imputed Income Earned on Owner-Occupied Houses

The Department of Statistics estimates the imputed net operating surplus earned on owner-occupied houses in preparing the national accounts.

Gross rents are derived from the 1986 Census of Population and Dwellings. Housing Corporation, other government department and local authority rents are not taken into account. Gross rents are projected forward by the ownership of dwellings sub-group of the producers (outputs) price index which in turn reflects the results of a survey of rents undertaken for the purposes of preparing the consumers price index. They are also adjusted to reflect the change in the number of dwellings which are estimated to be occupied by owners.

Housing-related expenditures, such as maintenance, insurance and water rates, are derived from the Household Expenditure and Income Survey. Consumption of capital (depreciation) is estimated at 2 percent of the amount of dwellings. Net insurance claims provide an estimate of the amount of unplanned obsolescence.

Based on the above general approach, the Department's estimate of net operating surplus for 1988/89 and 1989/90 (in current dollars) is as follows:

Owner-Occupied Houses

Net Operating Surplus

1988/89 to 1989/90

1988/89 1989/90

$m $m

Gross output (rents) 5,650 6,140

Less Intermediate consumption (expenditure) 1,379 1,727

Consumption of fixed capital (depreciation) 342 394

Indirect taxes 635 674

_____ _____

Net operating surplus 3,294 3,345

_____ _____

Source: Department of Statistics.

To derive an estimate of the subsidy from not taxing owner-occupiers on the implicit rents which they earn, two further steps are required. First, an allowance for interest costs needs to be made. Secondly, the appropriate tax rate needs to be applied to the estimate of net profits (operating surplus less interest costs) from owning a house.

Interest payments are estimated in the Household Expenditure and Income Survey 1988/89 and 1989/90 to amount to $1,741 million and $1,975 million respectively.

The appropriate tax rate to apply is the average effective marginal tax rate faced by owner-occupiers. Those who are earning taxable income in excess of $30,875 a year are generally on a marginal income tax rate of 33 percent. Individuals earning labour income over $9,500 and up to $30,875 are on a marginal income tax rate of 28 percent. Marginal tax rates are higher for certain individuals in receipt of Guaranteed Retirement Income or facing the phase-out of Family Support. Individuals on income-related interest rates or receiving other assistance from the government which phases out as income increases would also be on higher effective marginal rates. Finally, individuals who are in a tax-loss position (e.g. self employed) could be on a lower effective marginal rate of tax. The overall average effective marginal tax rate is not known. Home owners tend to be on higher incomes than renters. It seems reasonable, therefore, to assume a rate of 33 percent.

On the above basis the net income implicitly earned by owner-occupiers and tax forgone on such income is estimated as follows.

Estimated Tax Forgone on Imputed Rents

1988/89 and 1989/90

1988/89 1989/90

$m $m

Net operating surplus 3,294 3,345

Less mortgage interest 1,741 1,975

_____ _____

Net imputed income 1,553 1,370

_____ _____

Tax (33 percent) 512 452

The above analysis suggests that the exemption from income tax of net rents on owner-occupied housing provided a subsidy of around $500 million in 1988/89 and $450 million in 1989/90.

The Department of Statistics estimates of gross rents and net operating surplus, which are calculated quarterly, incorporate an average of 834,800 and 851,362 owner-occupied houses and flats in 1988/89 and 1989/90 respectively. Based on Valuation New Zealand's report on average sale prices in the years to June 1989 and 1990, and assuming that the houses sold in those periods reflect fairly the location and quality of the total stock of houses, the stock would have an average market value of $102,436 and $114,043 a house in 1988/89 and 1989/90 respectively. This would value the stock at $85.5 billion and $97 billion in 1988/89 and 1989/90 respectively. Gross rents and net operating surplus (net rents) would amount to around 6.6 percent and 3.9 percent respectively of the value of the housing stock in 1988/89. In 1989/90 gross rents amounted to 6.3 percent of the value of the housing stock while net rents amounted to 3.4 percent of the stock.

While the valuation of the housing stock on the basis of Valuation New Zealand's data on the average selling price of houses may overstate its value, it is even more likely that gross and net rents are well under-stated by the Department of Statistics because market rents for the existing rental stock are unlikely to reflect accurately the rental value of New Zealand's owner-occupied houses. Such houses could be expected to command a gross return of up to 10 percent suggesting that the tax revenue forgone from exempting imputed rents from tax may be up to 1.6 times as high as that calculated above.

It should also be noted that if such rents were taxable, then the prices of houses, gross rents and expenditure on houses could be expected to adjust. In addition, the additional revenue from taxing imputed rents might well justify a small reduction in tax rates. These factors have not been taken into account in the above calculation.

BIBLIOGRAPHY

Akerlof, George A. (1970), "The Market for 'Lemons': Quality Uncertainty and the Market Mechanism", Quarterly Journal of Finance, LXXXIV, 3, pp. 487-500.

Austie, Roslyn, Findlay, Christopher, and Harper, Ian (1983), "The Impact of Inflation and Taxation on Tenure Choice and the Redistributive Effects of Home-Mortgage Interest Rate Regulation", Economic Record, June 1983, pp. 105-10.

Bailey, W.J. (1988), Current Housing Policy: An Assessment, A Discussion Paper prepared for the Economic Planning Advisory Council, September 1988.

Bolger, J. B., Richardson, Ruth, Birch, W. F. (1990), Economic and Social Initiative - December 1990: Statements to the House of Representatives, Government Printer, Wellington.

Bray, Howard (1986), "The FHA: Selling Off a Dream", Across the Board, June 1986, pp. 8-14.

Brooks, R. (1986), Inflation, Tax Rules and Housing in New Zealand, Discussion Paper G86/4, Reserve Bank of New Zealand.

Building Industry Commission, Discussion Papers, Nos 1 and 2 (1989), Building Industry Commission, Wellington.

Building Industry Commission, News Nos 1-10 (1987-90), Building Industry Commission, Wellington.

Building Industry Commission, Working Papers, Nos 1-6 (1988-89), Building Industry Commission, Wellington.

Building Industry Commission (1990), Reform of Building Controls, Volumes I and II, Building Industry Commission, Wellington.

Building Societies Association (NZ) Inc (1987a), Housing Policy, unpublished.

Building Societies Association (NZ) Inc (1987b), Review of Housing Policy: A Submission to the Royal Commission on Social Policy, October 1987.

Cole, Bill (1989), Housing Policy or “Hello, I’m from the Government and I’m Here to Help You”, Australian Institute for Public Policy, Perth and Acton ACT.

Culhane, Dennis and Fred, Marc (1987), Paths in Homelessness: A View from the Street, Paper presented at the International Symposium "Affordable Housing, from Participation to Self-Help, and Shelter for the Homeless" September 16-20 1987, Hamburg FRG.

Department of Statistics (1987), 1986 New Zealand Census of Population and Dwellings: National Summary, Series C Report 2, Department of Statistics, Wellington.

Department of Statistics (1989), Household Expenditure and Income Survey 1988-89, Department of Statistics, Wellington.

Department of Statistics (1990a), New Zealand Official 1990 Yearbook, Department of Statistics, Wellington.

Department of Statistics (1990b), "Household Expenditure and Income Survey: Year Ended 31 March 1990", Hot Off The Press, Catalogue No. 09.505, Wellington.

Easton, Brian (1983), Income Distribution in New Zealand, New Zealand Institute of Economic Research, Wellington.

Elliot, Nicholas (1989), Streets Ahead, Adam Smith Institute, London.

Evans, Alan (1988), No Room! No Room! The Costs of the British Town and Country Planning System, Institute of Economic Affairs, London.

Fletcher Residential Limited (undated), Towards a New Zealand Housing Policy.

Freebairn, J, Porter, Michael, and Walsh, Cliff (eds) (1988), National Priorities Project 1988, Spending and Taxing: Taking Stock, Allen and Unwin, Sydney, pp. 138-53.

Friedman, Joseph and Weinberg, Daniel H (eds) (1983), The Great Housing Experiment, Sage Publications, Beverley Hills/London/New Delhi.

Gruen, Fred H (1988), Some Economic and Social Aspects of Housing in Australia, A Discussion Paper prepared for the Economic Planning Advisory Council, unpublished.

Housing Corporation and New Zealand Housing Initiative (undated), A Fresh Look at Housing, Action Kit.

Housing Corporation of New Zealand (1985), Report of the Housing Corporation of New Zealand: For the Year Ended 31 March 1985, Government Printer, Wellington.

Housing Corporation of New Zealand (1986), Report of the Housing Corporation of New Zealand: For the Year Ended 31 March 1986, Government Printer, Wellington.

Housing Corporation of New Zealand (1987), Report of the Housing Corporation of New Zealand: For the Year Ended 31 March 1987, Government Printer, Wellington.

Housing Corporation of New Zealand (1988a), Report of the Housing Corporation of New Zealand: For the Year Ended 31 March 1988, Government Printer, Wellington.

Housing Corporation of New Zealand (1988b), The Final Submission of the Housing Corporation of New Zealand to the Royal Commission on Social Policy, Housing Corporation of New Zealand, Wellington.

Housing Corporation of New Zealand (1989), Report of the Housing Corporation of New Zealand: For the Year Ended 31 March 1989, Government Printer, Wellington.

Housing Corporation of New Zealand (1990a), Report of the Housing Corporation of New Zealand: For the Year Ended 30 June 1990, Government Printer, Wellington.

Housing Corporation of New Zealand (1990b), Housing Corporation Rental Application Pointing System: Guidelines for Housing Corporation Staff and Housing Allocation Committees, Housing Corporation of New Zealand, Wellington.

Hudson, Edward A. (1989), Housing Corporation of New Zealand, Paper presented to Mortgage Lenders Seminar, 3 October 1989, Wellington.

McLintock (ed.) (1966), An Encyclopaedia of New Zealand, Government Printer, Wellington.

MGICA (1990), Housing Corporation Mortgage Guarantee Scheme, letter to New Zealand Business Roundtable, 10 August 1990.

Minford, Patrick, Peel, Michael, and Ashton, Paul (1987), The Housing Morass: Regulation, Immobility and Unemployment, Institute of International Affairs, London.

Ministry of Women's Affairs (1989), Women's Guide to Housing, Ministry of Women's Affairs, Wellington.

New Zealand Business Roundtable (1987a), Town and Country Planning: Toward a Framework for Public Policy, New Zealand Business Roundtable, Wellington.

New Zealand Business Roundtable (1987b), Comment on the Review of the Town and Country Planning Act 1977 by A. Hearn QC, New Zealand Business Roundtable, Wellington.

New Zealand Business Roundtable (1988), State Owned Enterprises: Issues of Ownership and Regulation, New Zealand Business Roundtable, Wellington.

New Zealand Business Roundtable (1989), Disaster Insurance Policy: A Submission to the Associate Minister of Finance, New Zealand Business Roundtable, Wellington.

New Zealand Business Roundtable (1990a), Disaster Insurance Bill: Submission to the Commerce and Marketing Select Committee of the House of Representatives, New Zealand Business Roundtable, Wellington.

New Zealand Business Roundtable (1990b), Resource Management Bill: Submission to the Select Committee of the House of Representatives, New Zealand Business Roundtable, Wellington.

New Zealand Housing Initiative, Forum, issues from July 1988 to September 1989, New Zealand Housing Initiative, Wellington.

New Zealand Housing Initiative (undated), Criteria for Economic and Environmentally Acceptable Residential Development, New Zealand Housing Initiative, Wellington.

New Zealand Housing Initiative (1989), Presentation to Planners’ Summer School, 20 February 1989.

New Zealand Institute of Economic Research (1991), Quarterly Predictions: March 1991, New Zealand Institute of Economic Research, Wellington.

Reserve Bank of New Zealand (1991), Weekly Statistical Release: For the Week Ending 11 April 1991, Reserve Bank of New Zealand, Wellington.

Rosen, Harvey S. (1985), "Housing Subsidies: Effects on Housing Decisions, Efficiency, and Equity" in Auerbach, Alan J. and Feldstein, Martin (eds) Handbook of Public Economics, 1, North-Holland, Amsterdam.

Royal Commission on Social Policy (1988a), The April Report Volume III Part Two, Future Directions, Associated Papers, Royal Commission on Social Policy, Wellington.

Royal Commission on Social Policy (1988b), The April Report, Volume IV, Social Perspectives, Royal Commission on Social Policy, Wellington.

Savage, John, Kerr, Susie, and Toplis, Stephen (1989), The Private Residential Rental Market in New Zealand, New Zealand Institute of Economic Research, Wellington.

Searle, J. and Scoular, P. (1983), Review of Planning and Building Controls, Wellington.

Sorensen, Anthony (1989), "The False Promise of Urban Consolidation", Policy, Winter 1989, pp. 19-21.

Stiglitz, Joseph E. and Weiss, Andrew (1981), "Credit Rationing in Markets with Imperfect Information", American Economic Review, 71, 3, pp. 393-410.

The Treasury (1984), Economic Management, The Treasury, Wellington.

Thomas, David C. Homelessness: A Private Trouble of a Public Issue, unpublished.

Thurow, Lester C. (1976), "Cash Versus In-Kind Transfers", American Economic Review, 64, 2, pp. 190-195.

Tucker, William, "Where do the Homeless Come From?" (1987), National Review, September 25, 1987, pp. 32-43.

Valentine, Tom (1989), "Home Ownership - Australian Dream or Australian Disease", Policy, Winter 1989, pp. 16-18.

Williams, Ross A. (1983), "Ownership of Dwellings and Personal Wealth", Australian Economic Review, 2nd Quarter 1983, pp. 55-62.

Williams, Ross A. (1984), Housing Policy in Australia, Research Paper No 113, Department of Economics, University of Melbourne.

Yates, J. (1982), "Tenure Choice and the Distribution of Income: Implications for Housing Policy", The Australian Quarterly, Autumn, 1982, pp. 63-75.

In addition to the above references, brochures and leaflets produced by the Housing Corporation of New Zealand and the Department of Social Welfare on housing were used as sources of information on assistance programmes.