In this article, we highlight the fact that low interest rates, while increasing demand for real estate, appear to be insufficient to create real estate booms, given the evidence observed in the United States over the past twenty-five years.
A simple logic
A mechanism often highlighted when discussing the influence of interest rates on demand for real estate is as follows: low interest rates (due in particular to an overly lax central bank policy) reduce the cost of credit for households, which then demand more real estate, pushing prices up. This logic seems to fit well with what has recently been observed with the real estate boom in the United States. For example, 3-month interest rates on US Treasury bills fell (from 5% to 1% between 2001 and 2004) and real estate prices soared (the Case-Shiller real estate price index rose by around 34% over the same period). However, over a longer period, the correlation between interest rates and real estate prices is much less obvious. According to the analysis by Glaeser, Gottlieb, and Gyourko (2011), lower interest rates can only explain one-fifth of the rise in real estate prices between 1996 and 2006, or about 10% of the appreciation over the period. This is certainly not insignificant, but it is far from the massive bubble that we observed. Does this mean that monetary policy had little impact on the US real estate bubble? Not necessarily, as we will see below. But first, let’s revisit the previous mechanism and try to understand why it may be less important quantitatively than we tend to believe.
Let’s take the simplest case of an individual (or household) who wants to buy a property. If this individual does not have the necessary funds to buy it without a loan, then an important factor in their decision will be the amount of the repayments they will have to make. If these are too high due to high interest rates, then they will prefer to rent rather than buy. A fall in interest rates automatically leads to a reduction in borrowing costs, which encourages some tenants to buy. However, at least three factors reduce the effect of a fall in interest rates on demand for real estate:
-First, we have been thinking in terms of fixed rates. However, abnormally low current interest rates have a relatively small impact when the loan is at a variable rate. In this case, the cost of the loan also depends on future interest rates.
– Household mobility for professional reasons or changes in family structure must also be taken into account. This can lead to the sale of the property. However, in an environment where real estate prices fluctuate, this can force households to sell when prices are low, which increases the real cost of buying a property.
– Finally, and this is undoubtedly the most important factor, if prices rise, this increases the cost of buying property. However, households that were unable to borrow due to low income or insufficient deposits are unlikely to benefit from lower interest rates.
This last point is particularly important when looking at the US case. Indeed, a key component of the real estate bubble is the expansion of mortgage lending to subprime households (low-income households with a history of late or missed payments).
In a 2009 study, Atif Mian and Amir Sufi suggest that it is indeed the easing of credit conditions that seems to explain most of the appreciation in real estate prices in the United States (and not the fact that individuals are seeing lower interest rates). The main finding of their study is that between 2002 and 2005, when broken down geographically, there was a negative correlation between income growth and credit growth. In regions with a high density of subprime households, income growth is low (and sometimes negative), while credit growth is high. Conversely, in areas where households are a priori more creditworthy, income growth is stronger and credit growth weaker. This phenomenon is unique in the last 25 years.
Furthermore, it is also these « subprime » areas that have seen the strongest appreciation in real estate prices. However, as noted above, one would expect the effect of a fall in interest rates to be relatively small on the demand for real estate among low-income households when prices are rising sharply. These households are, in fact, more constrained and therefore , in principle, less sensitive to a fall in interest rates. Yet the opposite has been observed.
Furthermore, during the period 1990-1994, there was a decline in interest rates comparable in magnitude to that of 2001-2004 (even though rates in 2004 were lower than in 1994), but there was no real estate boom at that time, and while real estate credit increased, it increased more in « prime » areas than in « subprime » areas. We therefore have two periods close together in time, with interest rates falling by the same proportion, and two completely different observations, whereas if the mechanism described above played a predominant role, we would expect two similar observations.
Does monetary policy therefore have no significant impact?
Saying that the mechanism described above plays a more marginal role than we think does not mean that monetary policy did not play an important role. In fact, we have only mentioned one mechanism, while others may also be at play. In particular, abundant liquidity may push financial agents to seek high returns. If there are mechanisms (such as securitization, which seems to have played a major role) that can generate high short-term returns through mortgage loans to households with low credit ratings, then a lax monetary policy can amplify the dynamics leading to a real estate boom. However, it is important to always identify the mechanisms that play a significant role from those that play a more marginal role.
References
Atif Mian and Amir Sufi, “The consequences of mortgage credit expansion: evidence from the US Mortgage default crisis,” Quarterly Journal of Economics, 2009
Edward L. Glaeser, Joshua D. Gottlieb, and Joseph Gyourko, “Can cheap credit explain the housing boom?”, National Bureau of Economic Research, 2011
