The Unassuming Economist

A new IMF Working Paper by Tigran Poghosyan “(…) use a novel dataset on effective property tax rates in U.S. states and metropolitan statistical areas (MSAs) over the 2005–2014 period to analyze the relationship between property tax rates and house price volatility. (…) [Poghosyan] find that property tax rates have a negative impact on house price volatility. The impact is causal, with increases in property tax rates leading to a reduction in house price volatility. The results are robust to different measures of house price volatility, estimation methodologies, and additional controls for housing demand and supply. The outcomes of the analysis have important policy implications and suggest that property taxation could be used as an important tool to dampen house price volatility.”

Please note that IMF Working Papers do not represent the official views of the IMF.

From the Global Housing Watch Newsletter: October 2016

How do we measure housing affordability? What does the data tell us about the United States? What explains the different patterns? What is the short term outlook? What can policymakers do? These are some of the questions that Svenja Gudell tackles in the November issue of the Global Housing Watch Newsletter. Gudell is the Chief Economist of Zillow and a leading expert on the U.S. housing market.

Svenja Gudell presenting housing market data (Photo credit: Adam Cohn)

Hites Ahir: Zillow has a lot of great data on the U.S. housing market and it is available to the public. So how do you track housing affordability?

Svenja Gudell: Zillow’s research frames housing affordability as the percent of someone’s monthly income spent on a mortgage or rental payment. For renters and buyers alike, we assume they make their area’s median annual household income and will be making the area’s median rent or mortgage payment. For home buyers, we make a few basic additional assumptions: that they are making a 20 percent down payment on the median-valued home in their area, and that they are financing their purchase with a conventional, 30-year, fixed-rate mortgage at prevailing mortgage rates as published by Freddie Mac. We publish this data assuming a resident is earning their area’s annual median household income and paying for median housing costs. In addition to considering the median, we also dig a little deeper to look at the tails of the distribution. So for a resident making their area’s median bottom-third household income (the median of the bottom third of the income distribution in that area), we match that buyer with either the mortgage payment of the median bottom-tier home – often a less expensive or entry level home – or the bottom-tier median rental payment. We do the same for the top third of the market, where residents making their area’s median top-third household income, are matched with the median top-third home or rental payment. We calculate all these affordability stats at the metro and city level.

We approach affordability in this way because we feel it provides an intuitive sense for consumers of the actual costs of housing to their household budget each month. It’s easy to communicate, for example, that renting the median home in the Seattle metro will consume 32 percent of a resident’s typical salary each month. More traditional measures of affordability, like price-to-income ratios or price-to-rent ratios, typically only produce a non-specific number that doesn’t tell the consumer very much. For example, the current price-to-income ratio in the Seattle metro is 5.2. In addition, this also doesn’t include the impact a low mortgage rate has on spending power.

Hites Ahir: What do the data show for the national and local level?

Svenja Gudell: In a general sense, buying a home currently is much more affordable than renting a home. Nationwide, buying the typical U.S. home on the median U.S. household income will consume about 14 percent of your monthly pay. Renting the typical home will consume 30 percent of income. Compared to historic norms, buying a home today is much more affordable than it was in prior decades: from 1985 through 2000, buying a home consumed, on average, about 21 percent of a buyer’s income. For renters, the story is reversed – renting today is much less affordable than it was in the pre-bubble years, when the typical rental consumed about 26 percent of income.

The reason buying a home is so much more affordable today comes down to today’s historically low mortgage interest rates, which help keep monthly mortgage costs low even as home values rise. Today, prevailing mortgage rates are hovering near or below 4 percent, far less than the 8 percent or higher rates that prevailed throughout much of the 1980s and 1990s. Mortgage interest rates will need to rise to roughly 7 percent before buying a home becomes less affordable than it was in the pre-bubble years (given forecasted home values over the next year), giving the overall U.S. housing market lots of headroom before mortgage affordability becomes a widespread concern. Renters, of course, can’t take advantage of low interest rates to help finance their rent, and so as rents have risen, the share of income needed to afford them has also gone up. Rental affordability problems are already emerging in many markets, and look set to only get worse as rents keep rising.

But while affordability looks pretty good at a national level (at least for home buyers) in a handful of very popular, pricey markets, the share of income necessary to afford a typical mortgage is already exceeding historic norms – even with historically low mortgage rates. In the Los Angeles metro, for example, buyers should currently expect to spend 38 percent of their income on a mortgage, up from 35 percent historically. If and when mortgage interest rates rise to 5 percent – a decent jump, but not outside the realm of possibility in the mid-term – Angelenos looking to buy a home should expect to spend almost half (46 percent) of their income on a mortgage. Renters in the L.A. area are already spending almost half their incomes on a typical rental payment (48 percent), and this share will only rise as rents keep going up at a faster pace than incomes. We see similar stories in other hot California markets, including San Diego and the Bay Area. In the long-run, these kinds of trends aren’t sustainable, and either incomes will need to grow very substantially, or housing costs will eventually need to level off or even come down a bit, in order to keep these communities affordable to typical residents.

Hites Ahir: If we look at housing affordability by income groups, what do the data show?

Svenja Gudell: It’s a lot easier for home buyers and renters at the top of the income distribution than it is for those at the bottom. Currently, those making the median, bottom-third U.S. household income and looking to buy the median, bottom-third/entry-level U.S. home should expect to pay roughly twice the share of their income toward a mortgage as those making the typical top-third income and buying a top-third home. In very expensive areas, like the San Francisco Bay Area, lower-income buyers should expect to pay almost 70 percent of their income toward a mortgage on an entry-level home. Essentially, in an area like San Francisco, it’s virtually impossible to afford housing costs on a low income and also afford all the rest of life’s necessities.

For lower-income renters, the situation is even more dire. Again in the Bay Area, a lower-income renter looking to rent even a less-expensive, bottom-third rental home will need to pay more than 86 percent of their income on rent – more than three times as much as a higher-income peer renting a more expensive, top-third home. Realistically, lower-income renters need to find multiple roommates to help shoulder this kind of burden; or try to work longer hours or find a second (or third) job to make ends meet.

Hites Ahir: What explains these patterns?

Svenja Gudell: Housing affordability is essentially defined by two basic factors: income and housing costs. If income rises at roughly the same pace as housing costs, then affordability doesn’t change much. If income growth exceeds growth in housing costs, then affordability can be expected to improve, in a general sense. If incomes stay flat or fall, but housing costs rise, then housing affordability will suffer. (Assuming relatively stable mortgage rates on the mortgage affordability side. As mortgage rates rise, affordability will also worsen, all else equal, as housing costs will rise.)

Over the past decade-plus, incomes at the upper end of the distribution have risen strongly, but incomes for the lowest earners have barely budged. At the same time, housing costs have risen fairly steadily over the past few years – especially rents, which largely did not see the big drop during the recession that home values did, and instead kept chugging upward. Exacerbating the problem at the lower-end of the market is the fact that bottom-tier home values are increasing at roughly double the pace of top-tier home values. This is largely attributable to low supply of homes for sale overall, especially at the bottom end of the market, and very high demand for those homes that are available. Investors, cash buyers and “regular” buyers alike are all competing for these homes, pushing prices up – which in turn contributes to mounting affordability issues.

Hites Ahir: In a recent article, you looked into the link between the housing bust and inequality. What did you find?

Svenja Gudell: The housing bust disproportionately impacted homeowners in lower-tier, less expensive homes – the share of entry-level homes foreclosed upon since 2006 is roughly three times the share of higher-end homes foreclosed upon over the same time. Most of these homes were foreclosed upon simply because their owners had less means to absorb some of the financial shocks of the recession, like a lost job. And those that were able to hang onto their homes often slid into negative equity, in which their home was worth less than the outstanding mortgage balance on it, making it virtually impossible to sell without a complicated, lengthy and costly short sale or mortgage modification. Once in negative equity – sometimes very deep – many of these homeowners simply gave up and let their home fall into foreclosure rather than keeping up with a mortgage and throwing good money after bad, exacerbating the foreclosure problem.

But in the years following the recession, the value of foreclosed homes climbed dramatically. Investors and cash buyers scooped them up to turn them into rentals, along with natural demand from first-time buyers and recent renters. Throughout the entire recovery, foreclosed homes showed greater annual appreciation than homes in general, peaking at 12.4 percent in January 2014, and falling to 6.8 percent by April 2016. Overall median U.S. home values, over the same time, reached a high of 7.9 percent annual growth in April 2014, with growth slowing to a pace of 4.9 percent by April 2016. Homeowners that were foreclosed upon, then, missed out on this opportunity to see large gains in their personal wealth as their home values grew. Their wealthier peers, able to purchase these homes at the bottom of the market and convert them into rentals, or homeowners that were able to hang on to their homes during the recession, only saw their wealth grow.

There is no small amount of irony in the fact that, after foreclosure, many former homeowners were prohibited from buying again with a mortgage for seven years, and so millions were forced to rent the exact same kind of homes they had owned only a few years prior. What’s more, these homeowners exchanged the relative stability and predictability of a monthly mortgage payment for the instability of rent.

And there’s still more salt to throw on the wound with the benefit of hindsight. It’s likely that millions of hardworking Americans found ways to hold on to their homes through the first few years of the recession, only to be foreclosed upon later – which actually turned out worse for them than simply giving the home up in the early years. A homeowner who foreclosed on a home in 2007 would have theoretically been able to buy again in 2014, and may have realized some of the gains in housing of the past few years. But a homeowner that held out desperately only to finally succumb to foreclosure in 2010 or 2011, won’t be able to buy again until 2017 or 2018.

Hites Ahir: In the short run, will housing affordability deteriorate or improve?

Svenja Gudell: We got some encouraging news recently that incomes last year rose at roughly the same pace as housing costs, a strong year of growth after years of stagnant or even falling wages. At a minimum, this kind of growth should at least mean that housing affordability maintains its current status quo, and doesn’t get worse. It’s good news, but one strong year of income growth isn’t enough to make up for several more years of low/no growth, coupled with very steady growth in housing costs. For housing affordability to meaningfully improve for most Americans – particularly renters – incomes will need to show solid growth for several more years to come. In addition, slowing home value appreciation and rental appreciation will help give incomes a chance to catch up.

Hites Ahir: What can policymakers do to address housing affordability issues? Are there any success stories that we can learn from?

Svenja Gudell: Solving housing affordability issues from a federal level is tricky, since so much of what determines housing affordability and the ability to build/maintain affordable housing is determined at a local level. In a general sense, policies which continue to promote job and wage growth will – obviously – help. Cutting down some of the local and regional barriers to creating affordable housing, like zoning constraints, permitting timelines and effective transit from areas with affordable housing to areas with abundant jobs will also help. Places where its relatively easy to build new housing – like Houston, for example – have succeeded in keeping housing affordable in a general sense, and there are lessons to be learned from their successes. Of course, not every market has as much physical room to grow as Houston or other wide-open areas in the middle of the country, so there is no “one size fits all” approach.

“The Central Bank of Ireland’s analysis of systemic vulnerabilities is sophisticated and timely. The Central Bank of Ireland has the power to request data directly from regulated entities, and also has powers to require information from unregulated entities under the Central Bank Acts. The Central Bank of Ireland also has powers to change the levels and regulatory perimeter of macroprudential instruments under national law, such as the LTV and LTI limits. There is a dedicated division (Financial Stability Division) that leads systemic risk analysis and macroprudential policy discussions. The biannual Macro-Financial Review (MFR) covers well the stability of individual sectors and property markets. There is, however, still room for further improvement, in particular as to filling data gaps. First, information on domestic and cross-border bilateral liability positions of banks and non-bank financial institutions is still incomplete in places. Second, detailed information on important elements of commercial real estate market activities is lacking. Third, balance sheet data for non-financial corporations is not fully available. Fourth, the absence of a comprehensive credit register precludes the Central Bank of Ireland from connecting credit information of borrowers across financial institutions in Ireland. Moreover, the Macro-Financial Review can usefully cover financial interconnectedness among sectors, as well as within each sector”, according to an IMF report on Ireland.

“On the real estate side, Portugal has 5.9 million housing units with an estimated value of €300 billion. House prices rose 9.9 percent between 2013Q1 and 2015Q4. Accordingly, the estimated value of the real estate assets owned by households increased by €25 billion, while household savings declined by €5 billion”, notes IMF report on Portugal.

From Global Housing Watch Newsletter: September 2016

In this issue of the Global Housing Watch newsletter, Shlomo Angel and Achilles Kallergis spoke with Hites Ahir about their work on global housing affordability. Shlomo Angel is a Professor of City Planning and Director of Urban Expansion at the Marron Institute at NYU. Achilles Kallergis is a Research Scholar in the NYU Urban Expansion program.

Shlomo Angel (left) and Achilles Kallergis (right)

Hites Ahir: The Land and Housing Survey in the UN Sample of Cities was recently completed. One of the aims of the survey is to provide a comparative analysis regarding housing affordability in a global sample of 200 cities. What is the main finding from this survey?

Shlomo Angel and Achilles Kallergis: Based on preliminary results from an analysis of 170 cities from the 200-city sample, the survey’s main finding is that cities across regions do face serious affordability challenges when it comes to housing. The median house-price-to-income ratio was 4.8 and the median rent-to-income ratio was 30 percent, values that are above what is generally considered affordable (house-price-to-annual household income ratio of 3.0, and rent-to-monthly household income ratio of 25 percent or less).

Figure 1: Housing Affordability in the Sample of Cities

Hites Ahir: To study housing affordability, you divide the housing sector between informal and formal housing. You then divide formal housing between public and private housing. And finally, private housing between multi-family and single-family housing. What are the shares for each type of housing at the global level?

Shlomo Angel and Achilles Kallergis: One of the most challenging tasks for our city-based surveyors was to identify different segments of housing and their respective shares in the overall housing sector. Certainly, the housing sector in each city is much more granular, yet, for the sake of comparability, we decided to adopt these broad categories. In terms of the global estimated shares, the survey results indicate that in the formal private sector, approximately 38 percent of housing is in multi-family buildings and approximately 34 percent in single-family housing. Public and informal housing in our global sample represent approximately equal shares of 14 percent.

Hites Ahir: How do these shares affect housing affordability?

Shlomo Angel and Achilles Kallergis: The different shares and segments of the overall housing sector play an important role in determining affordability. In many ways, the presence of an informal housing sector and public housing sector, which are the most affordable sectors across our city sample, acts as a cushion for low-income residents who cannot afford a dwelling in the formal private sector. If we omit the informal housing and public housing sectors, housing affordability is further exacerbated, particularly in cities that informal housing and public housing represent a significant share of the housing stock. To give an idea, excluding informal and public housing, the global median house-price-to-income ratio increases from 4.8 to 6.1, and the median rent-to-income ratio increases from 30 percent to 35 percent.

Now, there seems to be an interesting interplay between the formal and informal housing market, which based on survey data it would be worth exploring further. For instance, Brueckner and Selod offer an interesting theory based on the idea that informal and formal residents compete for land within a city and that the informal market “squeezes” the formal market, and therefore affects affordability in the formal sector. From a policy perspective, this type of interplay is important, as it implies that formalizing squatter settlements could have beneficial effects for both informal and formal residents.

Reference: Brueckner, J. and Selod, H. 2009. A Theory of Urban Squatting and Land-Tenure Formalization in Developing Countries. American Economic Journal: Economic Policy 2009, 1:1, 28–51

Hites Ahir: What does the survey results tell us about developing countries?

Shlomo Angel and Achilles Kallergis: One of the major messages of the survey is that affordability challenges are currently present in the growing cities of developing countries. A median house-price-to-income ratio of 4.8 is found in both developing and developed countries. Informal housing, which occupies approximately 18% of the housing sector in developing countries, is the most and in many cases the only affordable sector of the housing market. What is further worrying is that new formal private sector construction is often too expensive for median households, with the median house-price-to-income ratio for formal private multi-family buildings being 5.3.

For instance, in Kilamba, Angola, a satellite city in the outskirts of Luanda, a typical housing unit costs 17 times the median household income. The current unaffordability in developing countries has serious implications once we consider that between 2015 and 2050, 95% or 2.25 billion of the world’s urban population increase is expected to take place in developing countries. Unless, the housing question is addressed, unaffordability in developing countries will further exacerbate in the future.

Figure 2: The numbers on top of each bar represent annual incomes required for the purchase of an US$120,000 3-bedroom apartment in Kilamba, Angola, for each population decile (from low-income to high income).

Hites Ahir: And what about developed countries?

Shlomo Angel and Achilles Kallergis: Knowledge on housing affordability in developed countries is far more advanced and in many ways the survey results confirm previous studies on housing affordability. However, differences are present given that our definition of cities based on built-up space do not always coincide with administrative boundaries of urban areas. Median house-price-to-income ratio for developed countries is 4.8. In terms of shares of the housing sector, the bulk of housing units (90 percent) is in formal private housing; about one-tenth is in the more affordable public housing while informal housing represents only 1 percent. It is worth noting the importance of these shares as both private multi-family and private single-family housing are the least affordable segments of the housing market.

Hites Ahir: How does the housing affordability picture look like by geographic region?

Shlomo Angel and Achilles Kallergis: As expected, there is variation in terms of housing affordability by region with the most affordable regions being Southeast Asia, sub-Saharan Africa and land rich countries (United States, Canada, Australia) and the least affordable being Western Asia and North Africa, South and Central Asia, East Asia and the Pacific, and Latin America and the Caribbean. An interesting metric developed for the purposes of the survey, is to look at housing affordability of the formal private sector –what we call Median Affordability or simply, the ability of the median household in a given city to acquire or rent a housing unit in the formal private housing sector. Based on this metric, in all geographic regions the house price-to-income ratio is higher than the house price-to-income by Occupant Affordability, an affordability metric which includes the informal and public housing sectors. The differences are exacerbated in the regions where the informal sector and public housing sector constitute higher shares of the housing stock. In Sub-Saharan Africa, for example, the house price-to-income ratio measured by Median Affordability more than doubles.

Figure 3: Occupant and Median Affordability by Geographic Region

Hites Ahir: Could you provide two examples, where housing is affordable in one city and it is not affordable in another city? What could explain this divergence?

Shlomo Angel and Achilles Kallergis: In the Bangkok metropolitan area, housing is largely affordable with households contributing 3.3 yearly incomes to acquire a housing unit. In Beijing, China, housing is generally unaffordable with the occupant households contributing 12.1 yearly incomes to acquire a housing unit. While these differences are significant it is difficult to explain this divergence by simply relying on the survey data. In other words, affordability is a complex, idiosyncratic matter explained through a variety of reasons most often pertaining to city population growth, land use management and land use and housing regulations. One of our future tasks is to see how other metrics of the Monitoring Global Urban Expansion program such as the growth of the urban extent, the quality of urban layouts, and the land use and regulations regime in the 200-city sample relate to the affordability of housing. For instance, an interesting question arises as to the specific effects of land containment, through regulations such as the imposition of urban growth boundaries, greenbelts etc. to housing prices.

Hites Ahir: What are the key messages for policymakers?

Shlomo Angel and Achilles Kallergis: The central message from the preliminary survey results indicates a global housing affordability crisis that the formal private housing market alone is failing to confront. This affordability crisis cannot be addressed unless it considers all segments of the housing stock—including public housing and informal housing that are now the only housing solutions affordable to large segments of the urban population—while creating conditions for the private housing sector to reach further down-market. But beyond the changes in housing policies, and given the challenges ahead, there is a need to consider housing policy within a broader urban perspective. Housing is a key aspect of the urbanization process as it determines the spatial arrangements that regulate the cities productive structures and their ability to generate inclusive growth. For housing to be adequate, ample, affordable, and accessible to urban labor markets, land on the urban periphery must remain in plentiful supply and well connected by arterial roads; land and housing regulations must be made more realistic and responsive; property rights in housing must be better organized; and housing finance services must expand their reach.

Hites Ahir: What’s next for your research?

Shlomo Angel and Achilles Kallergis: The results from the survey represent a baseline for further monitoring housing affordability. We expect to continue this monitoring exercise in order to get better insights on housing conditions across the world. Further on, using the 200-city sample as a research vehicle, a sample that is representative of the universe of cities (all areas that in 2010 had a population of 100,000 or more) we hope to further explore critical aspects of the development of cities by adding layers of data. In this context, we are currently considering a survey on transportation that maps both formal and informal transportation networks. Finally, we are planning for a study of water quality across our 200 cities, through the collection and analysis of water samples from several areas in each one of the cities.