سیاست های پولی و رونق قیمت منطقه ای در سوئد

This study measures the heterogeneous effects of monetary policy on regional house prices in Sweden 1991–2002. We use a multivariate persistent shock metric to examine the impact of short-term interest rate on the property market. We also segregate the influence of interest rate shocks and influence of local shocks to capture their respective effects on regional price boom. We find significant regional effect of monetary policy on housing markets. Interest rate effects dominate the influence of local price innovations in the core economic regions in Sweden. We further discuss the monetary transmission channels and highlight political implications.

Over the last decade, the rapid and sustained escalation of home prices in Sweden has attracted a great deal of attention. From year 1991 through year 2002, house prices in Sweden increased by 1.5% per year in real level and 3.3% per year nominally. This large increase in house price has been associated with increases in household disposable income and with declining interest rates.3 According to the central bank of Sweden (Riksbank), from the beginning of 1990s, both short-term interest rate (6 months treasury bills) and long-term interest rate (5-year government bond) decreased dramatically in Sweden. During 1991–2002, nominal treasure bills decreased from 12.65% to 3.89%; while nominal 5-year government bond decreased from 10.52% to 4% over the same period. This stimulated household debt growing at a considerable high rate. The loans have been obtained largely to finance house purchase, in which more than 85% are secured with real estate. Consequently, the amount of debt has grown relative to disposable income significantly, the debt ratio to disposable income in year 2002 is around 100%, and it arrived at 145% by the end of year 2006. However, at the regional level, the rise in house prices has been very uneven across the nation. Regional prices roughly follow national cyclical trends, but with large differences in amplitude and some differences in timing (Fig. 1). The same price level applies to many regions in northern and southern Sweden, with the largest city in each metropolitan region tending towards a higher price level. The price of housing in Stockholm is three times higher than the national average. In regions of small population like Norrbotten (in far north Sweden), house prices are only one-fifth of the average national level (Yang & Turner, 2004a).Against this background, we consider two related research questions in this paper: (1) Should we expect monetary policy to have a homogeneous effect on housing prices across regions? (2) To what extent are recent regional prices boom related to changes in the level of interest rates? These two questions are critical to our understanding of the sources of risk for regional price volatility, and they have important implications for monetary policy. Monetary policy is focused in this study also because policy makers have paid increasing attention to the housing sector. Swedish central bank in early 2006 decided to increase rate despite justified with an explicit reference to rising household debt and housing prices.4 House prices play a special role in the conduct of monetary policy. The property market is important for monetary policy, not only because housing is a critical channel of monetary transmission and a leading indicator in the business cycle, but also because changes in house prices can have important wealth effect on consumption. Recent real estate prices volatility has raised concerns that housing could be a major determinant of macroeconomic volatility (Bernanke and Gertler, 2001 and Zhu, 2003). It tends to be a key factor in assessing the balance of risks to output and inflation. In order to formulate the appropriate policy responses, it is critical to identify the sources of prices volatility and the effects of monetary policy on housing. In this study, we address this issue in a context of regional responses. This is important to enhance our understanding of the impact of monetary policy because natural heterogeneous housing market raises the importance of regional dimension in housing study. However such a research is still very limited. The studies on spatial variation in house prices have been focused on the explanation of housing outcome from both demand and supply sides (e.g. Abraham and Hendershott, 1993, Abraham and Hendershott, 1996, Ozanne and Thibodeau, 1983 and Reichert, 1990). Variables such as interest rate, income, population and supply costs are indicated to be fundamental factors causing regional various in prices. Capozza, Hendershott, and Mack (2004) study the dynamics of housing price mean reversion and responses to income, population and construction or supply across 62 metro areas. They find heterogeneous responses of regional prices to these variables depend on information cost, supply costs and expectations cost. Hwang and Quigley (2006) explain regional house outcomes from national and regional economic conditions and specify the important role of vacancy in determining house prices. Goodman and Thibodeau (2007) analyze recent price appreciation in housing price in MSAs using a long-run simulation model. They find that the expected rate of appreciation in house prices is very sensitive to the assumed supply elasticity and they find significant spatial supply elasticity across regions. Glaeser, Gyourko, and Saksand (2004) andBarker (2004) suggest that supply regulation play an important role in regional price variety. Perhaps the closest study to this one is Negro and Otrok (2005) who study the effects of monetary policy on regional housing price in American. They explain the heterogeneity across states in the current house prices by the different exposures to national shocks. They find significant regional difference in the response of house price to monetary policy measured by Federal Funds rate. But monetary policy shocks are small relative to the size of recent price boom in their study. Although similar to their study, we attempt to understand the role of monetary policy in heterogeneous regional price boom; we also put efforts to gain insight into the identification of regional effect of monetary policy on house prices. Thus, following the monetary transformation channels, we discuss the possible factors that are important in explaining regional effect from both demand and supply sides. Political implications of regional monetary policy are also highlighted. Besides, we use a different persistence shock measure generally following Cochrane (1988) in this study. In this model, the interest rates itself proxies for monetary policy. We measure long-run responses of house prices to a unit shock of the interest rate. This is referred to as the persistence of interest rate shocks in the model. This model further separates the interest rate effects and orthogonal local effects on regional prices, measuring the relative impact of each. In this way, the model allows us gain insights into the driving forces of price fluctuations at the regional level. The paper is organized in the following way. Section 2 presents previous research that addresses monetary policy and housing market. In Section 3, we model regional effects of interest rate shocks in the Swedish property market, using a persistent shock method. In Section 4, we analyze the transmission mechanism of interest rate changes in the property market, indicating the potential causes of different regional responses to interest rate changes. We also highlight the political implications of our founding. Section 5 concludes.

In this study, we use the technology of shock persistence to measure the impact of interest rate shocks on regional house prices and we further segregate the influence of interest rate shocks and the influence of local shocks to capture their respective effects on recent regional price boom in Sweden. We find regional differences in the effects of interest rates shocks in Sweden and, in particular, we find that monetary policy contribute significantly to the price boom in the three biggest cities of Sweden, Stockholm, Göteborg and Malmö. We also try to understand the factors causing the regional effect of monetary policy along the monetary transformation channel and highlight the political implications of such a study. The science of explaining regional effects of monetary policy is still in the early stage of development. Further analysis needs to be done using both macro factors and micro information. It would be also worth examining the regional effect of monetary policy over the whole real estate cycle, and put insights into asymmetric effects of monetary policy at different phases of business cycle at the regional level.