When Covid hit and millions were suddenly forced to work from home, many predicted that a silver lining might be a mass exodus to the suburbs and countryside, thereby helping to tame buoyant home prices in cities. They couldn’t have been more wrong.
Fuelled by ultra-low interest rates, a savings glut and a yearning for bigger spaces, homes prices have surged in most major economies, outpacing income growth. Rate hikes this year have cooled the rally somewhat, but it is feared that by 2025 some 1.6 billion people will lack adequate housing.
National and local governments in Europe and the United States have introduced rent regulation in major urban centres including Berlin, Paris and Dublin with varying degrees of success. But the measures are often divisive, pitting tenants against landlords and investors and, in Sweden, even unseating a prime minister.
Given what is at stake, economists have scrutinised rent stabilisation and control measures for their impact. Studies show that these measures deter landlords and investors from entering the market, reducing housing supply and paradoxically increasing prices and rentals. Experts therefore tend to recommend supply-side policies, like the construction of high-rise apartments to increase availability of affordable housing.
Focus on well-being instead of housing supply
However, these studies tend to focus on how rent policies distort the market. Moreover, the prevailing solution of increasing housing supply benefits the rich as well as everybody else, whereas the poor would benefit more from targeted measures, especially in cities with high levels of inequality.
What if, instead of their impact on the housing market, rent policy measures are evaluated for their impact on a more fundamental metric: well-being? For this reason, my co-authors* and I created a model to simulate and assess housing affordability policies in any given locale, ranging from rent stabilisation (RS) – which is a cap on annual rentals hikes – to housing vouchers.
Using real-life data from the New York metropolitan area, we found that the policies improve average well-being, measured in terms of annual consumption of goods and housing, largely by guaranteeing lower-income residents a roof over their heads. The net welfare gains generated by affordable housing measures are substantial enough to outweigh the misallocations in labour and housing markets observed in previous research.
Our model is novel for factoring in income precarity and risk. In other words, if people have unstable incomes or are at chronic risk of losing their jobs – which is unfortunately common in large cities – then affordable housing policies that help them keep their homes are valuable. This is especially so given that most people are risk-averse, i.e. they dislike having to change their housing consumption like moving to a smaller apartment.
Rent stabilisation and the city
Cities draw people from all walks of life with job and educational opportunities as well as an array of lifestyle and cultural amenities. To understand the effects of affordable housing policies on diverse urban populations, my co-authors and I built our model to capture essential features of the housing market: house prices, rents, construction, labour supply and output, income, wealth inequality, and the location decisions of households. With the right inputs, the model can be used to study any city.
We calibrated our model to the New York metropolitan area and conducted five experiments. In the first, we increased the number of RS homes i.e. housing for which rent generally cannot rise above a few percentage points each year. In real life, RS covers as much as a third of apartments in New York, with another one-third transacting at market rates and the remaining third occupied by owners. Our model shows that if RS housing is increased by 50 percent (by square footage), it will increase low-income households’ access to stable housing. In turn, New Yorkers’ average welfare will increase by a substantial 0.91 percent.
But this policy is not without costs. RS housing units are typically older, smaller and often allocated without means testing, which means they sometimes go to people who can in fact afford to live in non-RS, bigger apartments. Suppressing rent also inevitably weakens incentives to construct and maintain housing while driving up prices of market-rent units. As a result, some professionals and highly skilled workers may be forced to live in the cheaper suburbs and waste time on commuting to the city centre.
Next, we tested two measures that might offset the above-mentioned misallocations. The first is a more targeted approach that income-tests every RS tenant periodically to ensure more RS units go to people who in fact need them. However, we found that income testing reduces incentive to work since earning more may disqualify one from RS housing. And, because some long-term RS tenants are replaced by (needier) new tenants, the average subsidy is smaller as new RS tenants get much smaller rent discounts. All told, this policy delivered substantial welfare gains of 0.66 percent.
The next policy simulation relocates all RS housing from the urban core to the suburbs. Unsurprisingly, the result is gentrification of the city centre with fewer but higher-income residents, larger apartments and increased home ownership. Viewed in a positive light, this means more high-skilled, high-productivity households can live in the heart of New York, contributing more to the city’s output than lower-productivity households.
The flip side is the marginalisation of low-income households to the suburbs. Not only does this lead to segregation of the haves and have-nots, a large part of the population also miss out on amenities – museums, theatres and the like – available only in the city centre. In all, average welfare increases by a comparatively modest 0.25 percent.
Our fourth policy experiment increases the amount of housing available to everyone – not just low-income residents – by easing land-use or height restrictions on housing construction in the city centre. Rents fall across the board as a result. However, because the measure generates only a modest improvement in the plight of low-income households, the aggregate welfare of the population rises by a mere 0.11 percent.
Voting with their feet
Interestingly, giving an additional US$800 million in housing vouchers, essentially cash transfers, to low-income households produces no welfare gain in our model. Any benefit generated is offset by higher taxes introduced by the authorities to fund the expenditure. This causes some high-income residents to move away, necessitating even higher taxes to finance the same expenditure, in turn driving out more wealthy residents, and so on. The outflow, meanwhile, triggers a drop in the housing stock, resulting in higher rents.
Our model can be a useful tool for policy makers investigating the benefits and costs of various rent stabilisation measures for their cities. While there is no one-size-fits-all solution, one thing is for sure: Rent stabilisation policies can be a net positive for average well-being in unequal cities such as New York and Seoul after the associated risks and efficiencies are accounted for (though perhaps less so in more equal cities with vast social safety nets). By assuring financially precarious residents of housing stability over the long term, such policies ultimately give more than what they take away.