Some thoughts on affordable housing I

This blog has argued in multiple posts (here, here, here, here, here, here, here, here, here, and here) that affordable housing availability is one of the biggest challenges to sustainable urban growth in developing countries. I'm inclined to place urban housing affordability gap right up there with poor student learning outcomes as the two biggest and most complex development challenges today. 

The conventional wisdom has been that if restrictive zoning regulations are removed, ease of getting building and other permissions simplified, and taxes on construction lowered, it will increase the stock of housing of all kinds including affordable housing. This, in turn, will lead to lowering of house prices - "richer renters trade up into new luxe units, starting a chain of move-ins and move-outs that lower prices for modest homes". This can be called the "trickle-down" theory of housing affordability. 

Instead, I'm inclined to believe that a meaningful dent in the housing affordability issue in the medium-term has to involve an increased supply of large volumes of public (or heavily subsidised) housing. And the aforementioned, market side measures mentioned above if implemented well should take a long time to start showing impacts. In the coming weeks, from articles that have been published in recent days, I'll blog about the hugely impressive Viennese rented social housing, Gemeindebauten, which highlights the former, and the example of Washington DC, which highlights the latter. 

In a blog post, Noah Smith points to several papers that find increased supply of market rate housing results in lower rents - every 10% increase in housing stock, rents decrease 1% and sales prices also decrease within 500 feet"; "rents fall by 2% for parcels within 100 m of new construction, and renters' risk of being displaced to a lower-income neighbourhood falls by 17%"; market rate housing construction reduces rents for low-income housing market specifically; "adding new homes moderates price increases and therefore makes housing more affordable to low and moderate-income families" etc. He also links to papers from Germany and Finland that confirm the downward pressure on prices due to increased supply of market-rate housing. A detailed set of links here. Further, at a macro-level of a city, there's new evidence that cities that have built more housing have seen smaller increases in average housing prices. 

But theory and several data points aside, in reality, there are several reasons to argue that this trickle-down theory for affordable housing may not always play out as expected. This could especially be so in developing countries and their cities. 

There is an important reason. The housing market suffers from an important distortion, arising from certain kinds of people viewing housing less as a place to live but as an investment class. Francesca Mari points to a McKinsey report which found that in the US 68% of growth in asset values over the last four decades came from real estate, and that one in four home sales in 2022 was to someone with no intention of living in it. 

Accordingly, if left to the dynamics of the market, developers will respond by catering to the highest margin (and therefore most profitable) demand. Given that investment residential properties are higher in value and offer much higher margins, it's inevitable that the market will respond to this demand. As the City Lab writes

The market leads developers to build luxury housing on scarce and sought-after property to maximize the return on their investment.

And this will be at the cost of the cheaper and smaller margin affordable housing units. In these circumstances, the solution to affordable housing is to increase the supply of housing stock that people want to use for living, as against those that people want to use for investment. 

We need to step back here a little bit. The starting conditions of the cities under consideration matter. For example, how segmented is the housing market, how do the prices in the different segments compare, what's the likely profit differential between higher-end and marginal housing, what's the marginal demand for higher end housing, how does it compare with the supply, how do the starting prices for each housing segment compare with the annual incomes of different population groups etc. Differences in each can generate entirely different outcomes.  

And on each of these aforementioned factors, I'm inclined to argue that there is a significant difference in the nature of the housing markets in developing countries from those in developed countries. Some general observations:

1. In stratified markets (housing is extremely segmented), the aggregate increase in supply may not only not leave everyone better off, but may also leave the vast majority worse-off. So, an increase in the supply of higher value houses would displace the fixed land and scarce capital that would otherwise have been available for affordable housing. And the per capita land footprint occupied by high-value individual housing is orders of magnitude higher than that occupied by affordable housing (unit-wise or individual-wise).  

Noah Smith has a nice metaphor to explain this

It makes perfect sense to worry about rain when you see people opening their umbrellas, even though the umbrellas don’t actually cause the rain. The reason is that when you see a bunch of people opening their umbrellas, it means they’ve probably checked the weather forecast. Umbrellas aren’t a cause of rain, but they are a signal of rain. A harbinger. An omen. Market-rate housing development is similar. If you’re a working-class person, and you see a big new shiny glass apartment tower going up a block away, it makes perfect sense to be afraid that rents are about to rise. The apartment tower tells you that A) your city is becoming more of an employment hub for yuppie types, and B) for some reason, the yuppie types have decided that your neighborhood is a good one to live in. But that doesn’t mean you think the new apartment tower is the cause of the gentrification.

2. Increased supply of investment housing, especially in blighted and renewal areas, often also leads to gentrification of the area. At the extensive margin, this crowds-out those less well-off, while at the intensive margin, it increases the prices of the existing affordable housing units. This is a macro/regional level reflection of the micro/housing-level crowding out that happens with housing as an investment.  

Further, in large countries like India and their densely populated cities like Mumbai, there is another important dynamic at play. As I have blogged on several occasions, the higher income market segment is itself so large and its unmet demand very high that incremental supply will be heavily skewed towards this segment. This will ensure that the top market segment draws most of the housing investment and limits the growth of affordable housing supply, and that too for a very long time. 

3. As a model, if left to the market dynamics, the trajectory of deregulation of zoning regulations, ease of permissions, and renewal of blighted areas will attract developers looking to maximise their returns. At the unit level, this leads to a much higher supply investment properties than affordable housing. At the level of localities, this leads to gentrification and displacement of poorer people. 

In fact, in the largest Western cities, there has been a net decline in affordable housing in recent years. Sample this from New York City (via this report),

Between 2017 and 2021, New York City lost almost 100,000 units that had rented for less than $1,500 per month, while it added 107,000 units that rent for at least $2,300 per month... In Manhattan, for example, the median effective rent in April 2022 was $3,870, more than 38 percent higher than a year before and the highest level ever recorded.

4. The lesson here is that if governments merely deregulate, ease permissions, and release government lands, on their own these measures will not result in higher supply of affordable housing. Worse still, they could reduce the incremental supply of affordable housing from the marginal unit of land. These measures should be complemented with measures that directly target creation of affordable housing units.  

5. Apart from directly building public housing, the instruments governments could use to target affordable housing could include higher FAR for such developments, lower building permission and other development charges, even lower property taxes for a certain period, fast-tracked clearances, and concessional capital by way of interest subvention. Perhaps confine these benefits to limited profit entities, as the Austrian example shows. 

6. There's a strong case for higher levels of taxation on vacant houses, so as to deter this trend. But it runs into the problem of identification and monitoring. 

In this context, in an effort to discourage property speculation, Singapore has just announced higher taxes by foreigners to purchase residential property in the country. Further, permanent residents and citizens buying their second residential property will also pay more in the form of an additional buyer's stamp duty (ABSD).