021 946 2480 info@rode.co.za

Inclusionary housing: The risks and opportunities


Erwin Rode contemplates the law of unintended consequences when inclusionary programmes are introduced in SA and comes up with some practical suggestions

The City of Johannesburg has proposed a policy in terms of which developers will be compelled to provide a minimum of 20% ‘low’ and ‘low-middle’ income housing units in each new residential development.

The City defines ‘low-middle’ as household incomes of less than R7 000 per month. The proposal stipulates that rentals, including levies (but excluding utility bills, like electricity), may not exceed R2 100 per month in 2018 rands. The ceiling of R2 100 is based on 30% of a household income of R7 000. (City of Johannesburg, 2018)

The proposed programme is designed to provide rental accommodation at an affordable price. Two types of management are promoted, viz.

  • Social housing. This is no problem, as these institutions are geared to manage low-income tenants.
  • Private ownership with a capped rental (as above). In the case of sectional title developments, the inclusionary housing units must be owned, managed and rented out by the body corporate. This is a non-starter, as the management of a body corporate is complicated, and it would put a tremendous burden on the trustees, apart from the fact that it is a challenge to find competent trustees among the owners in complexes where the average selling price is, as a rule of thumb, below R1 million.

Generally, the purpose of an inclusionary housing programme is (a) to increase the number of units built for low-income categories and (b) to promote the integration of socio-economic classes.

This is nothing new, in fact in the USA more than 500 jurisdictions have implemented inclusionary housing policies, and these can be found in 27 states, as well as Washington DC. The most productive programme in the USA ‒ Montgomery County in Maryland ‒ goes back to 1974 (more about Montgomery County later).

By design, inclusionary programmes link the production of low-spec housing to market-related housing production. If market-related housing construction stalls, so does the construction of low-spec units (Sturtevant, 2016:8).

There are many variations on such a policy programme, some of which are:

  • Obligatory vs. voluntary. In the USA about 83% of programmes are obligatory. The City of Johannesburg proposes to make it mandatory.
  • The types of incentives to offset the cost to the developer. It goes without saying that when the programme is voluntary, the incentives should be adequate. If the programme is compulsory, as proposed by the City of Johannesburg, the incentives should also be adequate, otherwise it could inhibit the development of all classes of housing, which would eventually push up the prices of existing housing stock.¹ In the literature, examples of incentives are:
  • Density bonuses (conditional rezoning to allow more height and more bulk²)
  • Reduced parking requirements
  • Rebate on contribution to the cost of engineering services
  • Rebate on assessment rates payable to the municipality
  • Fee waivers
  • Expedited planning approvals.

These just about exhaust the incentives a municipality can offer without incurring serious costs to itself. Density bonuses and reduced parking requirements suggest that high-rise inner cities are prime candidates for promoting inclusivity. On the other hand, a study in Montgomery County (Maryland) concluded that density bonuses are not effective in promoting low-spec housing in areas that were already zoned for high-rise buildings. The reason for this is that after a certain height, land costs become an increasingly smaller proportion of overall development costs (Sturtevant, 2016:9). Thus, for the City of Johannesburg, the low-hanging fruit may be areas currently zoned for low bulk, and commanding low land prices, but bordering on high-rise precincts, which can be rezoned on condition of inclusivity.

Fee waivers and expedited planning approvals are nice gestures, but would not improve the viability of a development by much, if at all.

  • A buyout option, allowing developers to pay an in-lieu fee to an affordable-housing fund (e.g. a social housing institution³). This is not part of the proposal by the City of Johannesburg.
  • An off-site option, allowing developers to develop low-spec housing in a different locality. However, the problem with this option is that developers could choose inexpensive land on the outskirts of the city, thereby reinforcing the so-called apartheid spatial planning. A better option would be to channel off-site developments to degraded inner-city areas bordering the CBD, where conditional rezoning to a higher bulk would add value. However, this is not part of the proposal of the City of Johannesburg.
  • The policy could apply jurisdiction wide (as proposed by the City of Johannesburg) or it could apply only to a defined precinct (e.g. a CBD or an Urban Development Zone).
  • Small developments are exempted. The City of Johannesburg proposes to exempt developments of fewer than 10 dwelling units.

The question now is to what degree these policies in the USA have (a) led – via depressing new construction ‒ to an acceleration in house-price inflation with respect to the existing stock of housing, and (b) to what degree these programmes have stimulated the construction of low-spec housing.

These issues were addressed in a meta-analysis⁴ by the Center for Housing Policy (the research division of the National Housing Conference) in the USA. For various reasons, the outcomes of the studies are inconclusive with respect to the effect on housing starts and, therefore, price inflation of existing stock. Thus, we must rely on common sense and economic theory.

Regarding additions to low-spec stock, the most productive ‘inclusionary’ jurisdiction in the USA ‒ Montgomery County in Maryland ‒ could show in total an increase of 13 246 low-spec units between 1974 and 2010. This equals an increase in the stock of low-spec housing units by 358 units per year over a period of 36 years. To put this gain in perspective, the total stock of housing units (all prices) in 2010 was 375 905, so we can say the gain of 13 246 units boosted the total stock of units by altogether 3,6% after 37 years. This gain must be weighed up against the potential discouragement of market-related housing construction and, as a consequence, accelerated house price inflation.


In sum, my comments on the proposal by the City of Johannesburg are as follows:

A programme to enforce inclusionary housing is risky as nobody knows what the unintended consequences will be ‒ the most obvious risk being a strike by residential developers. To them, the biggest risk is that their sales tempo in new developments would be retarded to the extent that the development becomes unviable. From the developers’ point of view, the obvious part solution to this problem is to keep the price gradient between the various price classes within a development shallow. For example, mix low-spec units with houses that do not cost more than, say, R500 000. However, the downside to such a reaction by the private sector would be (a) that the profit margins on low-priced houses are wafer thin and (b) that no more houses or residential units of more than R1 million would be constructed. Using economic theory and common sense, this would lead to spiralling house prices in these categories.

In the USA, such policies have contributed very little to additional low-spec housing stock. Thus, given the risk outlined under (1), a municipality should approach such a programme with the utmost sensitivity and care.

  1. To mitigate the risk, the City of Johannesburg could consider applying its policy not to the whole municipal area, but only to selected, degraded centrally located areas where land prices are currently low, but where value could be added through allowing more bulk together with other incentives like rebates. Urban Development Zones come to mind.
  2. If proposal (3) is accepted, then the obligatory nature of the programme would mitigate most of the risk to the greater residential market. If proposal (3) is not accepted, then the City should make the programme voluntary. In either instance, the incentives should be substantial enough to offset the costs to the developer.
  3. The proposal that, in the case of sectional title schemes, the body corporate should own, manage and rent out the low-spec units in the scheme, is a recipe for disaster. Trustees of such schemes are typically volunteers who work pro bono, thus to expect them to be saddled with the hassles of conduct issues, not to mention bad debts, is unrealistic. As a rule of thumb, in schemes in which the market-related units cost less than R1 million, it is very hard to find competent and willing trustees. However, this comment does not apply to social housing institutions, as they are geared to manage tenants in low-spec units.

City of Johannesburg. (2018). City of Johannesburg inclusionary housing: incentives, regulations and mechanisms (draft for public comment). Johannesburg.
Sturtevant, Lisa A. (2016). Separating fact from fiction to design effective inclusionary housing programs. Center for Housing Policy.