During the years leading to the global crisis, the US and Germany were the dominant growth poles in the Americas and Europe, respectively (ADD CITE). Their position reflected their growth performance and their dominant size. Both countries were characterised by contrasting patterns of public policies towards home ownership and education – the US put greater public commitment to subsiding home ownership, whereas Germany put much greater public commitment towards education and vocational training.
The US has prioritised housing ownership, promoted by preferential tax treatment of mortgages and home ownership, subsidising quasi-public institutions designed to deepen the ownership of real estate.1 In the US, there is a lower and declining public commitment towards subsidising quality education (pre-school, primary, secondary, vocational, and tertiary levels). In contrast, Germany does not provide public subsidies for home ownership, and has a much deeper residential rental market. Germany is also committed towards deeper support of low-cost quality education, including a well-resourced vocational education and training system, combining public and private funding, aiming at preparing blue collar workers for the challenges facing labour in the modern economy.2
The residential housing market, key trends, 1994-2006
Before the global financial crisis, the home ownership rate for the US trended upwards, from about 64% in 1994, to a peak of 69% in the mid-2000s, presently at 66%, and declining (USHOWN, St. Louis FED). In contrast, home ownership rates in Germany have been stable, at about 40%, and among the lowest in Europe.3 The patterns of the appreciation of real estate, exhibited by Figure 1, illustrate the huge gap in policy outcomes. During the lead up to the global financial crisis, 1994-2006, home prices in the US exhibited cumulative nominal appreciation of 115%, whereas in Germany there was a nominal depreciation of 4%. While the housing boom in the US cannot be attributed only to the preferential tax treatments of mortgages, the interaction of this policy with other contributing factors played a role in accounting for the remarkable US real estate boom-bust cycle.4 While the interest rate in Germany was at a comparable level to that of the US, most of the other factors that played a role in the US were absent in Germany, inducing a remarkable decline in the relative real price of real estate.
Figure 1. Housing index (1994=100)
Source: Aizenman and Jinjarak (2009).
Manufacturing, key trends, 1994-2006
The decline of the manufacturing share in GDP is a common feature of convergence and the development of the services sector, and has characterised the global economy in recent decades. Yet, countries differ substantially in their levels and trends of manufacturing-to-GDP, as is vividly illustrated in Figure 2. Before the crisis, the manufacturing-to-GDP share of the US was about half that of Germany. In the 15 years before the crisis – the post-reunification years – Germany’s manufacturing-to-GDP share was flat. In contrast, the US continued the sharp decline of its manufacturing-to-GDP during the 15 years before the crisis, by another 4% GDP share.
Figure 2. Manufacturing share in total value added
Source: OECD STAN
Interpretations and puzzles
The public policy in the US has provided preferential treatment to residential investment rather than to investment in human capital and vocational training, relative to the policy stance in Germany. The period of rapid real appreciation of real estate in the US coincided with the continuation of rapid decline in its manufacturing/GDP ratio. In contrast, Germany avoided bubbly real estate, while maintaining a stable manufacturing-to-GDP ratio at time of robust economic growth (Figure 3). Our work suggests a robust association across countries between rapid real appreciation of real estate and the decline in the manufacturing-to-GDP ratio in the years prior to the crisis.
Figure 3. GDP per capita (PPP)
Source: World Development Indicators
The ex-post evidence suggests that the policy mix chosen by the US did not work as intended. There is no clear evidence that subsidising home ownership instead of human capital and vocational education led to greater income equality or reduced poverty over time.5 The crisis of 2008-2009 revealed the downside risk of US policies in magnifying the exposure to bubbly real estate dynamics.
Observing these two policy approaches to home ownership versus human-capital and vocational education, does not necessarily suggest these are connected. We postulate, however, that there may be several ways in which this divergence between the US and Germany along these two dimensions is indeed linked (for more detail see Aizenman and Noy 2012).
The first possibility is plain-vanilla fiscal constraints. Fiscal resources are limited, and spending more than 5% of GDP on home ownership by means of lucrative tax exemption and subsidies precludes higher spending on education. However, the data on US education spending does not necessarily point to lower levels of spending, but rather to mediocre educational outcomes. US educational outcomes may be related not only to spending but to systemic difficulties in controlling quality in sprawling and disjointed systems. Quality control may require centralisation, and an effective stick-and-carrot policy.
Regressive redistribution subsidies
Another related possibility is that regressive redistribution subsidies for homeowners weaken the lobbying power of the educationally underserved. Higher home ownership (or a trend upward in rates and in prices) leads to a decreasing interest among policymakers in spending on education; since homeowners in appreciated property feel they can better afford privately funded education. There may also be other relevant dynamics. For example, homeowners may only consider the quality of their specific local school while renters, who are (maybe unwillingly) more mobile, care more about the general level of education in their region. Furthermore, since the subsidies associated with home ownership are highly regressive and (maybe) tilted toward the older generation, they increase the political clout of those who are less interested in equitable access to education. Thus, higher home ownership will cause less willingness to invest in education.
Misreading the meaning of higher home ownership
We also believe that policymakers may misread the meaning of higher home ownership. Policy desire to reduce income inequality or poverty (if it is indeed a policy aim) is perceived as successful when housing ownership indicators are trending up; thus removing the perceived need to use education as a leveller (and invest in particular in education at the lower end of the distribution for, for example, blue collar workers). This may help explain why there was bipartisan support for the Bush administration’s housing policies during the 2000s.6
Different policy preferences
More plainly, the difference in policy emphasis between Germany and the US may just reflect different policy preferences. The lower home ownership rates in Germany are a result of a stronger desire to assist the lowest-income households through a more extensive social housing programme. In this case, both lower ownership rates and the provision of high-quality technical training are both a result of these policy targets (lower poverty/reducing inequality). US policymakers did not share this aim, but were more interested in electoral support from the middle class.
We also note that during 1994-2006 Japan shows a German-style non-bubbly housing market, and a correspondingly flat manufacturing-to-GDP ratio, while the UK is very similar to the US in terms of real estate appreciation and the decline of manufacturing.7
What explains the persistence of questionable policy choices in the US?
While both parties in the US supported heavily subsidised home ownership, it is not obvious why an inefficient policy survives the test of time. If one should choose to subsidise only one sector by 6% of the GDP at time of scarcity of tax revenue, why is subsidising residential investment preferred to human capital and vocational training?8
The sustainability power of the US inefficient policy configuration may be explained by the dynamic effects of subsiding home ownership. Once a subsidy of residential investment was adopted, it provided immediate gains to house buyers and to the construction sector. Importantly, this policy generated a growing club of supporters, as scaling down the policy would lead to capital loses for homeowners and to the construction sector. In contrast, a policy of subsidising investment in human capital and vocational education may take a long time to generate the social and private rewards. The elimination of this policy may increase the quasi-rents of the older population, which has greater political clout relative to the young and the unborn.9
Aizenman, Joshua and Yothin Jinjarak (2009). "Current account patterns and national real estate markets," Journal of Urban Economics, 66(2): 75-89.
Aizenman, Joshua and Ilan Noy (2012) “The contrast of public policies between Germany and the US: Real estate versus human capital,” work in progress.
Bourassa, Steven C. and William G. Grigsby (2000) “Income tax concessions for owner‐occupied housing,” Housing Policy Debate, 11 (3): 521-546.
Hanushek, Eric A. (2010) "The Difference is Teacher Quality", in Karl Weber (ed.), Waiting for "Superman": How We Can Save America’s Failing Public Schools. New York: Public Affairs, 2010, pp. 81-100.
Mian, Atif R and Sufi, Amir (2009). “The Consequences of Mortgage Credit Expansion: Evidence from the US Mortgage Default Crisis", Quarterly Journal of Economics, 124:1449-1496.
Rajan, Raghuram G. (2010). Fault Lines: How Hidden Fractures Still Threaten the World Economy, Princeton Press.
Shiller, Robert (2000). Irrational Exuberance, Princeton University Press (2nd edition, 2005).
Voigtländer, Michael (2009). “Why is the German Homeownership Rate so Low?” Housing Studies, 24 (3): 355-372.
1 The Tax Policy report estimates the deductibility of mortgage interest on owner-occupied homes in 2012 to provide a tax subsidy of $609 billion. The exclusion of net imputed rental income provides another tax subsidy to real estate of $303 billion. Both subsides add up to about 6% of the US GDP, and are among the top items in the implicit subsidies provided by the US tax system. In contrast, the treatment of education expenses is close to being tax-neutral (see here and also see Bourassa and Grigsby 2000 for the history and analysis of housing ownership preferential treatment of the US tax code).
2 As with medical services, comparing education services across countries requires adjusting the real costs by quality measures, which are hard to assemble. The GDP share into public investment in education is similar in the US (about 4.8%) and Germany (about 4.6%). The GDP share into private investment in education in mid 2000s was higher in the US (about 1.6%) than in Germany (about 0.9%) (see here). Thus, the GDP share of total investment in education in the US exceeds that in Germany. Yet, the evidence suggests that the quality of the average outcome is higher in Germany than in the US. The 2009 PISA Comparison of countries’ performance shows that Germany significantly outperforms the US in Science and Math scores (520,513 in Germany versus 502,487 in the US. Germany’s performance is above the OECD average, while the US’s is statistically significantly below (see here). (See also Hanushek (2010) for analysis of the inefficiency of the education expenditure in the US.) While the US remains strong at the top end of the tertiary education, the public university system in Germany remains a bargain relative to the university system in the US in terms of affordability. Similar observations apply to vocational education and training.
3 Voigtländer (2009) points out that the comparatively low homeownership rate in Germany is due to four factors: “First, rental housing makes up a larger share of the market because of an extensive social housing sector. The high quality standard of social housing and the fact that private investors were included in the subsidisation scheme from the beginning laid the foundation for a large private rental housing market. Second, homeowners in Germany did not benefit from the same high subsidies as in countries such as Spain or the Netherlands. Third, the German rental housing market was not rendered inoperative by excessive interventions in rents, as was the case in countries such as Spain and the UK. Finally, German house prices remained stable over a long period of time.”
4 The Federal government during the 2000s pursued other policies whose aim was to increase home ownership rates; these were part of President Bush’s Ownership Society initiative. From a White House Fact Sheet (9/8/04): “The President believes that homeownership is the cornerstone of America's vibrant communities and benefits individual families by building stability and long-term financial security. In June 2002, President Bush issued America's Homeownership Challenge to the real estate and mortgage finance industries to encourage them to join the effort to close the gap that exists between the homeownership rates of minorities and non-minorities.” Related factors include the rapid increase in the loan to value of new mortgages; the bundling of mortgages that were resold in secondary markets, the changing the role of banks into commission seeking agents (Mian and Sufi 2009), the decline in interest rates and the proliferation of ‘teaser interest rates’ needed to extend houses affordability to lower quality borrowers, households growing presumption in the early 2000s that residential real appreciation is here to stay [ignoring Shiller’s (2000) warning]; and the deregulation drive propagated by the growing acceptance of Greenspan’s seductive assertion of “market-stabilising private regulatory forces.” (12 April 1997 speech, see here).
5 The Gini coefficient of income inequality of Germany has been below that of the US by about 10 points, and the gap between the two countries trended upwards during the years prior to the 2007-2008 crisis.
6 Republicans’ Ownership Society programme is at least in part aimed at making the lower-middle class (traditionally leaning democratic) friendly towards the importance of private assets and the protection of asset markets in general (see Rajan 2010 on the political economy of house ownership in the US in the pre-crisis years).
7 The US/UK dynamics are also consistent with the ‘Dutch disease’ type effects of the real exchange rate, where strong real estate and high rents in services like the financial sectors in US and UK before the crisis reduce competitiveness of manufacturing. On the other side, the housing markets dynamics are affected also by demographic developments, where Japan and Germany are aging faster than the US and UK. All these issues deserve more research.
8 Indeed, economic theory provides plausible arguments in favour of subsidising investment in human capital and vocational education instead of investment in bricks and mortar.
9 The sustainability of public policies towards investment in human capital and vocational education in Germany, and the smooth functioning of the rental market there may reflect the history of Germany from the Bismarck era. When the provision of public services is efficient enough, it generates a partnership where the voters support sizable involvement of the public sector in the formation of human capital, and keeping the sustainability of the rental market.