Synchronization of Housing Markets: How? What?

Keyword: 
Financial crisis
Topic: 
Financial Economics

There have been intensive discussions about the synchronized behavior of house prices and its implications for the real economy over the past decade. How synchronized are house prices across countries? And, what are the main shocks driving movements in global house prices? This column addresses these questions. House prices have become more synchronized over time. Global interest rate shocks tend to play a significant role in explaining movements in global house prices whereas global monetary policy shocks per se do not seem to be important. Interestingly, global uncertainty shocks account for a significant fraction of the variation in global house prices.

Researchers did not worry about the synchronization of house prices prior to the 2000s because there was a commonly held belief that these prices were mainly driven by domestic shocks as housing was seen as the ultimate non-tradable asset. The past decade has dramatically changed this perception as highly synchronized movements in housing markets appeared to play an important role especially during the global financial crisis.

House prices first increased unusually rapidly prior to the global financial crisis reaching in some cases levels not previously seen. They then collapsed over the period 2006-11 and have recently started to rebound in some of the advanced countries. The highly synchronized fluctuations in housing markets first coincided with a period of rapid economic growth, but then were followed by severe financial disruptions and deep recessions.

In light of these observations, this column addresses two specific questions to have a better understanding of fluctuations in global housing markets: First, how synchronized are housing cycles across countries? Second, what are the main shocks driving movements in global house prices?

How synchronized are house prices?

To answer this question, we study several measures of synchronization of house prices for 18 advanced OECD countries. These measures include basic correlations, concordance indices, and variance decompositions of factor models that allow us to examine the roles played by global and country-specific components in explaining national house prices. We study the evolution of house price movements over two different sub-periods: 1971:1-84:4 and 1985:1-2011:3. The second sub-period represents the “globalization period” in which there were dramatic increases in the volume of cross-border trade in both goods and assets.

The analysis of these indicators reveals a consistent narrative about the synchronization of house price fluctuations.

• House prices tend to move together but they exhibit less synchronization than most other real and financial variables.
• House prices have become more synchronized over time. The degree of concordance of housing cycles has increased from about 50 percent during the pre-globalization period to roughly 65 percent in the globalization period.
• The fraction of variance of house prices explained by the global house price factor has increased from about 20 percent during the pre-globalization period to 35 percent in the era of globalization. (Figures 1A, 1B, and 1C)

Cross-country correlations

 

Concordance of cycles across countries

Concordance of cycles across countries

Notes: Fist panel shows the average cross-country correlation, second panel shows the median concordance across countries, and third panel shows the average variance explained by the global factor, in respective time periods. The full sample covers the period of 1971:1-2011:3. The pre-globalization period is the sub-period of 1971:1-1984:4, and the globalization period is the sub-period of 1985:1-2011:3. Concordance is the fraction of time that two cycles are in the same phase. First, the concordance for each country pair is calculated, then the median for each variable over the sample is presented with average values in parenthesis.

What drives global house prices?

We use a variety of models to examine the importance of different types of shocks in explaining global house price fluctuations. Four major results about the sources of global house price fluctuations stand out from this analysis.

First, global interest rate shocks tend to have a significant effect on global house prices in most specifications when these shocks are identified using a standard recursive scheme (Figure 2). This finding has led some observers to argue that the use of monetary policy of low interest rates prior to the 2007-09 global financial crisis has been an important driver of house prices in advanced economies, especially in the United States. Shocks to global income, credit aggregates, and equity prices play a small role but innovations to house prices themselves account for about half of the variation in house price series. This result indicates that although a large fraction of house price movements is left unexplained, shocks to global interest rates do appear to play a major role. However, it is not clear what drives interest rate movements: are they market driven? Or are they simply determined by central banks?

Figure 2. Varience Decomposition of House Prices (Recursive)
(fraction of varience due to each shock, full sample, in percent, in 12 quarters)

Varience Decomposition of House Prices

Notes: This figure shows the proportion of forecast error variance of house prices explaned by the respective shocks for 12 quarters forecast horizons. The model includes output, house prices, interest rates, credit, and equity prices, respectively.

To shed light on this issue, we turn to an identification scheme that allows us to produce impulse responses compatible with standard theoretical predictions. In particular, we identify global monetary policy shocks using sign restrictions. This brings us our second finding: global monetary policy shocks per se do not appear to have a sizeable impact on global house price movements (Figure 3).

Figure 3. Varience Decomposition of Houses Prices (Sign Restrictions)
(fraction of varience due to each shock, globalization sample, in percent, in 12 quarters)

Varience Decomposition of Houses Prices

Notes: This figure shows the proportion of forecast error variance of house prices explained by respective shocks for 12 quarters forecast horizons. Uncertainty shocks are for the G-7 sample, all other shocks are for 18 advanced countries.

Third, neither global productivity nor global credit shocks appear to have a significant impact on house price movements. Productivity shocks tend to have a small impact on output, and shocks to output have only a modest influence on the growth rate of house prices. With respect to the minor role of credit shocks, we argue that credit markets often function well and shocks stemming from these markets are on average small. The implication is that shocks originating in credit markets have on average a small impact on house prices.

However, this result does not preclude the possibility that when there are sizeable movements in credit―during periods of credit booms or crunches, for example―credit market shocks can lead to significant fluctuations in house prices, as the global economy witnessed over the past decade. Innovations to house prices themselves account for about half of the variation in house price series. We view this as the fraction of movements in global house prices that we are unable to explain with the Factor-Augmented Vector Autoregressive (FAVAR) model.

Last, but not least, we find that uncertainty shocks tend to have a significant impact on global house price movements. Specifically, one third of the global house price cycles can be attributed to uncertainty shocks, which are measured by the volatility of equity returns, in the sample of G-7 countries. This is an economically large role for a single shock to explain, and hence an important empirical fact.

From a mechanical perspective, uncertainty series are highly correlated across countries implying that they have a significant common component. This component also appears to be important in explaining the common variation in house prices. Another interpretation of our finding is that when the volatility of stock returns increases, agents appear to shift their portfolio composition towards safer assets, such as housing. In addition, our findings also indicate that positive uncertainty shocks are associated with a decline in interest rates, perhaps reflecting a reaction from the monetary policy authority, which also plays a major role in driving global house price movements.

Housing Markets: No Longer Disjointed…

House prices in advanced economies tend to move together and the extent of synchronization of national housing markets has been increasing over time. Although both world interest rate and uncertainty shocks are important in explaining global house price fluctuations, global monetary policy shocks per se do not appear to play a major role. Understanding the sources of synchronized movements in global housing markets is a challenge and future work is needed.


The views expressed in this article are those of the authors and do not necessarily represent those of the IMF or Federal Reserve Bank of St Louis.

References

Hirata, Hideaki, M. Ayhan Kose, Christopher Otrok, and Marco E. Terrones, 2013, “Global House Price Fluctuations: Synchronization and Determinants,” IMF Working Paper No: 13/38

 

 

Share this