International Capital Flows, House Prices, and the Euro

Free Exchange’s weekly reading list includes an excellent recent paper by Jack Favilukis, David Kohn, Sydney C. Ludvigson, and Stijn Van Nieuwerburgh: “International Capital Flows and House Prices.” Lots of observers (including me) have noted the suspicious correlation between surges in international capital flows into certain countries in the early 2000s (e.g. the US, Ireland, Spain, Greece, Iceland, Australia) and simultaneous or near-simultaneous surges in house prices in those countries. This paper addresses the question of whether there is in fact a systematic relationship between capital flows into a country and house prices. Were the house price booms of the 2000s caused by international financial flows?

The answer provided by this paper is no, or at least not directly. When different possible macroeconomic explanations for changes in average national house prices are considered, it turns out that by far the most important factor is the ease of bank credit. In other words, rising house prices in the 2000s (as well as their subsequent fall) probably had much more to do with the willingness of banks to lend than any other factor. When banks are happy to lend money and they relax lending standards, house prices go up. When banks reverse course, house prices go down.

The importance of bank lending standards to the US housing bubble has been well documented and discussed, but this data suggests that the same may be true for a number of other countries as well. On the other hand, countries that did not experience a general relaxation in lending standards in the early 2000s did not experience house price booms. Once changing lending standards are taken into consideration, changes in international capital flows seem to have little additional explanatory power for house price changes.

This raises an obvious question: why did credit standards change in certain countries in the early 2000s? Bank lending standards are surely partly endogenous (as the paper discusses) – when banks expect house prices to continue rising, they are more willing to lend, which helps to push house prices higher. That sort of self-fulfilling logic is exactly why changes in house prices (first up and then down) were so extreme in the boom countries between 2002 and 2009. But this story doesn’t explain how the cycle got started in the first place in those countries.

For that, we need to look for some factors that can affect bank lending standards that are external to the housing market. Surely, general prospects for macroeconomic growth must play a role there, as well as overall risk tolerance. When a country seems to be headed for better economic times and risk tolerance grows, banks become generally more willing to lend. And that is where we come to the euro. (Were you wondering when I would bring that into the story?)

The peripheral euro countries benefited in specific tangible ways from adoption of the euro in 1999, not least from surges in international capital flows that reduced interest rates. Yet this research demonstrates that there is no direct connection between those capital flows and house price booms. So how is the euro involved?

This paper provides some evidence that in addition to truly exogenous changes in the supply of bank loans, expectations about future economic growth also have an impact on house prices: all else being equal, when growth prospects improve house prices go up. And more generally, bank lending standards depend heavily on their perception and tolerance of risk.

Now consider the likelihood that the adoption of the euro by the peripheral European countries (e.g. Spain, Ireland, and Greece) created expectations for higher growth (and lower interest rates) in those countries, and helped persuade banks to become less risk averse. House prices start to rise and banks become more willing to lend. House prices rise more. Banks respond by relaxing credit standards further. And the bubble begins to inflate.

Surges in capital flows don’t directly create house price bubbles. But this paper does help us understand a mechanism by which the adoption of the euro could have indirectly caused house price booms: by changing expectations and altering the perception of risk in the eurozone periphery, a self-reinforcing cycle of easier credit was sparked in those countries. That’s not all there is to it, of course – other factors surely must have also caused changes in risk aversion and bank lending standards in the housing bubble countries – but it does seem to be a likely piece of the puzzle for the peripheral eurozone.

Putting it all together, we now have a plausible contributing explanation for why almost all of peripheral Europe experienced a house price boom following the adoption of the euro, while the euro core (Germany, Austria, Benelux) missed it. It’s yet another way in which adoption of a common currency by economically dissimilar countries may have vastly important but completely unforeseen consequences.