UBC Theses and Dissertations
Quantification of the cost of alternative forms of housing market intervention in Canada Johnston, Kevin James
This thesis is concerned with the estimation of the cost involved in supporting minimum housing standards under alternative forms of government intervention. No attempt is made to rationalize what minimum housing standards should be. The intent is to highlight what the relative costs would be to support alternative housing standards, utilizing potentially alternative policy approaches. To rationalize the numerous policy/program alternatives available to governments, attention is focused on six policy alternatives that reflect varying degrees of "leakage" of the provided subsidy to the consumption of non-housing goods and services. These alternatives cover the continuum from pure cash (income) subsidies, wherein up to a 100% "leakage" to the consumption of non-housing goods and services may occur, to direct product intervention, wherein the recipients may be required to reduce the consumption of non-housing goods and services in order to improve their housing standards. The six options are classified by type either as a "cash" subsidy (where the use of the subsidy is not constrained) or a "direct product" subsidy (where the subsidy is used directly for the subsidization of capital and/or operating costs). Under the "cash" subsidy approaches two benchmark options are considered: 1. an "income" policy that involves the subsidization of incomes such that the recipients are free to choose consumption levels. This policy corresponds to pure income redistribution. 2. a "constrained cash" or "in kind" policy that involves the subsidization of incomes at a level dependent on either the level of housing consumption or expenditure on housing consumption. Such a policy covers programs comparable to rent certificates and general housing allowances. The options considered as "direct product" approaches are more arbitrary in nature, reflecting alternative benchmarks: 1. a "direct constrained" policy wherein recipients are forced to reduce their consumption of other goods and services to offset the required increase in expenditure on housing. This policy form is taken as the lower benchmark for the policy continuum, reflecting the potential impact of zoning and occupancy regulat ions. 2. a "direct quantity" policy wherein recipients are compensated for the cost of the additional quantity of housing consumed only. 3. a "direct expenditure" policy wherein recipients are subsidized in the amount of the total increase in expenditure on housing incurred. 4. a "direct price" policy wherein the subsidy is provided to buy down the unit price of housing. Such a policy covers mortgage market intervention and compensated rent control programs. The computation of these costs, for selected Canadian Metropolitan Areas, (CMAs), is undertaken using a simulation model based on a long run (ten year) regional housing market model developed by the Urban Institute, Washington DC, during the early 1970's. This model traces the interaction inherent in the supply and demand of housing in each of many separate, but highly inter-dependent housing submarkets. For this study the demand side of the model has been respecified to allow the incorporation of a tenure option, the use of after tax incomes and the direct estimation of the demand functions. The essence of the modified model is a microeconomic perspective on households and owners contracting for housing at prices and quantities determined in several submarkets. A selected CMA is represented in the form of several residential zones that reflect variations in the housing stock, income levels and distance from the Central Business District. On the demand side, model households are classified into one of ten groupings (reflecting age, family status and income earners) each having different preference schedules. These preference schedules are based on translog utility functions which compare housing consumption, after-tax income, leisure time and neighbourhood quality. On the supply side, a housing unit is characterized by the quantity of "housing services" (combination of size and quality) and the price per unit of service. Several different quality levels are distinguished to correspond to the many alternative quality submarkets in an actual housing market. Profit maximizing behaviour by the owners is assumed to imply linear supply curves (of varying slope for each type of housing unit) as an approximation of the price-quantity relations that govern the behaviour of owners over a ten year period. An unlimited volume of new construction is allowed at fixed unit prices, and in any size that provides a quantity of services greater than a defined minimum. The major inputs to the model are a CMA's set of supply and demand parameters, the 1971 decade-end demand profile and the 1961 decade-start market state. A model solution is an estimate of the market conditions at decade-end. The predicted market conditions are based on an assignment procedure which sets unit prices such that no household has incentives to relocate and owners have no incentive to provide an alternative quantity of "housing services". While the model is conceptually simple and assumes convenient forms for the household preferences and dwelling supply curves, it captures many aspects of housing markets important in distinguishing the impact of alternative subsidy policies. Policy alternatives may be introduced by simply adjusting the demand and/or supply functions to allow straightforward simulations of the necessary subsidy costs under alternative policy options. Eight CMAs were selected for the intervention cost study, reflecting a distribution with regard to geographical location, population and income growth during the 1961-71 decade. In implementing the model for a CMA the number of decade-end model households and dwellings is set at 100, reflecting a ratio of actual dwellings to model dwellings ranging from 287 to 3455 for the selected CMAs while the model parameters have been estimated alternatively from Census data or the 1974 Survey of Housing Units data. Certain parameters relating to neighbourhood externalities and the supply functions must be estimated by comparing model solutions with actual decadal performance. The computed costs of intervention derived from the simulations show a significant variation firstly in terms of the alternative forms of policy and secondly by CMA, reflecting the relative states of the existing housing stock. The most generous form of government support lies with an "income" policy wherein households are provided with sufficient income to consume the minimum quantity of housing by choice. The other end of the scale is represented by a "direct constrained" policy whereby households are forced to increase their housing standards, but the compensation level is only sufficient for them to achieve their prior level of overall satisfaction. The simulated difference in cost between these two alternatives is in the range of a factor of 10. That is, it could cost the government up to ten times more per annum to support housing standards via an "income" policy. The margin between the costs associated with the "income" policy and other alternative "direct product" oriented modes is in the range of a factor of 4 to 5. The results suggest that a general "direct product" oriented policy, tied directly to the change in the quantity of housing consumed (supplied) will cost less than one-third of the amount associated with a general "cash" policy. At present costs this margin corresponds to an additional amount in the region of six billion dollars per annum for Canada.
Item Citations and Data