VoxEU Column Financial Markets International Finance

Financial globalisation and securitisation in mortgage markets

Mortgage-backed securities have played a major role in the financial crisis and aren’t very popular as a result. This column documents macroeconomic benefits of these instruments, showing that economies with more developed markets for securitised mortgage debt share more consumption risk with other economies.

The securitisation of mortgage-related debt has played a major role in the emergence and proliferation of the current financial crisis (Brunnermeier 2009). Understandably, this has led to widespread about the usefulness of such instruments for allocating macroeconomic risk. It is now obvious that the repackaging of mortgage debt in mortgage-backed securities can have enormous aggregate costs, but so far no one has empirically documented the macroeconomic benefits of these instruments.

How should we measure such benefits? Some simple economic theory may help. One way to assess the usefulness of a financial innovation (such as a mortgage-backed security) is to ask whether it helps diversify risk. Now, at a macroeconomic level, the ultimate risk we all face is fluctuations in consumption; our incomes and wealth may go up in booms and down in recessions. But to what extent a recession really hurts is determined by whether we have to move out of our house, sell our car, or just cut down on the odd restaurant meal. Hence, whether people are able to maintain stable consumption over the business cycle is an important indicator of how well financial markets allow them to spread risk. And the extent to which an economy securitises mortgage debt may impact society's ability to diversify such consumption risk internationally.

This impact is potentially large. Mortgage markets are far less internationally integrated than, say, equity or bond markets, and residential real estate is largely domestically financed in most countries. Figure 1 plots the international correlation of stock markets against that of residential housing prices. For all of the 120 country pairs considered here, housing prices have much lower international correlations than stock markets.

Figure 1. Pairwise correlations of quarterly housing and equity returns


Note: The sample period is 1985Q1:2008Q1 for 16 industrialised countries.

Together with the fact that housing wealth is the single largest component of private net wealth in most countries, the figure suggests that fluctuations in the value of residential real estate (or of the debt collateralised on it) constitute a potentially important source of idiosyncratic risk from the perspective of the individual economy. In a recent paper (Hoffmann and Nitschka 2009), we argue that securitisation has helped diversify such risks internationally because it has made mortgage debt internationally tradable.

Securitisation and consumption smoothing

To measure the extent of securitisation in a country, we compiled information on the de jure introduction and use of securitised mortgage debt in a cross-section of 16 industrialised economies. Our qualitative measure does not primarily focus on the amount of outstanding securitised debt but on the ease at which this debt can be traded and the extent to which the issuer remains exposed to fluctuations in the value of the collateral. These legal differences should matter for risk sharing because they mean that securities collateralised by mortgage debt may effectively represent different assets when issued in different countries and under different jurisdictions. Our indicator allows for different degrees to which the secondary market for mortgage debt can be liberalised in a country. A zero indicates that the country has no such market at all, unity describes the countries with the most liberalised markets (such as the US) and intermediate values stand for intermediate forms of securitisation (such as the existence of markets for Pfandbriefe – a form of covered bonds used in Germany and Switzerland). As countries liberalise over time, our measure increases.

Figure 2 shows how the average de jure possibility for securitising mortgage debt increased. A jump in the line means that another country or group of countries introduces or liberalises the secondary market for mortgage debt. There is rich variation in the extent and timing of such liberalisation.

Figure 2. Trend in the qualitative degree of securitisation


Note: Country acronyms indicate which countries introduced mortgage securitisation at the respective date.

We exploit this variation for our empirical analysis. A flavour of our main results is presented in Table 1. To measure how well consumption is shielded from country-specific business cycle shocks, we run a panel regression of consumption growth rates on output growth rates in which we control for common time effects by removing world-wide fluctuations in consumption and output. The estimated coefficient tells us how strongly a typical idiosyncratic business cycle shock affects a country's consumption – it is a measure of unshared consumption risk. To gauge the impact of securitisation on this coefficient, at each point in time, the panel is split between the countries with the most liberalised and the least liberalised markets for mortgage debt. While the identity of these countries changes (due to liberalisations occurring which changes the ordering of countries), this procedure generates two synthetic representative groups of the most and the least liberalised markets, while at the same time controlling for other, potentially unobserved, country differences that also might affect risk sharing.

The results suggest that the more developed a country's markets for securitised mortgage debt, the more consumption risk it shares with other countries. While a one percentage point decline in a country's output (relative to the world average) leads to a 0.8 percentage point decline in consumption in countries that do not allow securitisation in mortgage markets, the decline is only 0.6 percentage points where secondary markets for mortgage debt are most developed.

We find that this result is significant and robust to a range of alternative specifications and, in particular, to controls for concurrent developments in world financial markets such as the secular growth in international cross holdings of other financial assets over the 1990s and 2000s that is also known to have increased international risk sharing.

Don’t neglect the benefits of securitisation

Hence, securitised mortgage debt is an important aspect of financial globalisation that makes people's consumption considerably more resilient to the ups and downs of the business cycle. One possible reason why securitising mortgage debt has such a strong effect on international risk sharing may be that markets for residential mortgages used to be some of the most internationally segmented parts of the financial system. Once banks were allowed to repackage and sell parts of their mortgage portfolio (often to foreign investors), they were able to continue to provide credit to consumers even in downturns, thus effectively providing consumption risk sharing to private households.

Indeed, we find evidence that supports the view that a stabilisation in bank loan supply has played a role in the transmission – countries with more developed markets for securitised mortgage debt have experienced higher growth in their credit to GDP ratios and this growth appears less sensitive to changes in interest rates. This international evidence is consistent with findings for the US that show that extending banks possibilities for diversification of their loan portfolios has measurable positive effects on credit market access for households and firms and, ultimately, on risk sharing. In related research, Hoffmann and Sherbakova (2008) argue that US state-level banking deregulation during the 1980s allowed banks to extend more credit to small firms in downturns, which in turn has stabilised interstate risk sharing. Loutskina and Strahan (2009) find that securitisation has weakened the link between bank financial conditions and mortgage loan supply.

The results surveyed here suggest some important policy lessons. Yes, securitisation has generated some severe moral hazard problems, such as reducing lenders' incentives to properly monitor credit quality and promoting opaqueness (Demyanik and van Hemert 2008; Mian and Sufi 2008). But in regulating these markets, policymakers should be very careful not to throw the baby out with the bathwater. Securitised mortgage debt also seems to have brought some measurable macroeconomic benefits in the form of better international consumption risk sharing. The regulatory challenge will be to improve securitisation in a way that does not jeopardise these benefits.

References

Brunnermeier (2009), "Deciphering the 2007-08 Liquidity and Credit Crunch", Journal of Economic Perspectives, 23(1), 77-100
Demyanyk, Yuliya S. and Van Hemert, Otto, Understanding the Subprime Mortgage Crisis (December 5, 2008), to appear in the Review of Financial Studies
Hoffmann and Nitschka (2009), “Securitisation of Mortgage Debt, Asset Prices and International Risk Sharing” Institute for Empirical Research in Economics Working paper no 376, University of Zurich
Hoffmann and Shcherbakova (2008), "Consumption Risk Sharing over the Business Cycle: The Role of Small Firms' Access to Credit Markets", Institute for Empirical Research in Economics Working Paper No. 363; CESifo Working Paper Series No. 2544
Loutskina and Strahan (2009), "Securitization and the Declining Impact of Bank Finance on Loan Supply: Evidence from Mortgage Originations", Journal of Finance 64(2), 861-889
Mian, A. and A. Sufi (2008). "The Consequences of Mortgage Credit Expansion: Evidence from the 2007 Mortgage Default Crisis", NBER working paper 13936.

4,514 Reads