What Causes Asset Price Bubble in an Emerging Economy? Some Empirical Evidence in the Housing Sector of India

Abstract

This study examines the dynamic causal relationships between house prices and their five determinants – real income, short-run real interest rates, real stock price index, real effective exchange rate, and real non-food bank credit – by using the quarterly data from 1996:Q1 to 2007:Q1 for India. Using the cointegration test and the vector error-correction model (VECM), the study finds that in the long run, real income significantly and positively influences the housing prices while real non-food bank credit adversely influences it. The variance decomposition results suggest that it is the shocks to the non-food bank credit that mainly explains the variability in housing prices, besides its own shocks being the most influential while other factors are not significant. This suggests that the role of credit availability as a supply side determinant cannot be underestimated in the dynamic behaviour of housing prices in emerging economies.

Keywords:

• Housing price
• stock price
• interest rate
• bank credit
• exchange rate

JEL CLASSIFICATIONS :

• R31
• G12
• E44
• F41

Acknowledgements

We are very much thankful to the participants of 11th annual national conference on ‘Money and Finance’, held at the Indira Gandhi Institute of Development Research, Mumbai (India) on 23–24 January 2009 for comments on an earlier version of this paper. We are especially grateful to Dr. Vijayamohanan Pillai, Associate Professor at the Centre for Development Studies, Trivandrum, India and the anonymous referees of the manuscript for their valuable inputs. Their helpful comments have greatly improved the paper. However, all remaining errors are ours.

Notes

¹The speculative impact of the stock market on the economy is quite different from the speculative impact of the housing market. The impact of the stock market on the economy is temporary in nature compared with the impact of housing market. The housing price bubble takes time to occur.

²The study has used the word ‘bubble’ in order to describe the housing price fluctuations and thereby its effect on construction sector activities, as there has been lots of variation in housing prices in recent years compared with prior to the 1990s, especially in the 1970s and 1980s. The Indian housing market is quite a new phenomenon as there was no market for it (housing was not being traded in the past, unlike the USA’s mature housing market). India’s housing market is an emerging and growing one. In India, it has only recently been seen that along with a rise in housing, there is also sharp rise in prices. This is due to the increasing demand compared with its supply. Once supply becomes excessive in relation to the income, one can expect a crash. The slump is observed during a period of global crisis, like other countries, although there is difference in the magnitude of the impact of the crisis across the economies or markets. This depends upon the opening up of the sector or exposure of the sector to the external sector.

³See the Abelson et al. (2005). Economic theory suggests that real rates are more important because the nominal component of interest costs should be offset by a nominal increase in housing price. The nominal interest rates can create a repayment problem in the early years for some borrowers and restrict their borrowing. An unregulated interest rate may reflect real borrowing costs better.

⁴Housing has a volume cycle, not a price cycle. Home prices are sticky downward. When there is a decline in price, house owners would not rush to sell the house, expecting that there would be a further decline in prices. Rather, they hold it at a lower price (Leamer, 2007).

⁵The Government of India (GOI, 2005) took the bold step to liberalize the external economy, the housing sector in particular. The Government has decided to allow FDI up to 100% under the automatic route in townships, housing, built-up infrastructure and construction development projects, which would include commercial premises, hotels, hospitals, recreational facilities and regional-level infrastructure, etc.

⁶The study also considers the 364-day treasury bill rate in place of the 91-day treasury bill rate for verification of results. However, it is subsequently seen that the results remain unchanged.

⁷The results are not reported here due to space constraints but can be made available upon request.

⁸In the Johansen Test, we can calculate the trace statistic, i.e. λ _i (0) = –T [ln (1–λ ₁) + ln (1–λ ₂) + ln (1–λ ₃). Where Tis the total number of observations less the lags and λ _i are the characteristics roots of the coefficient matrix of independent variables. Similarly, the same formula can be used in the calculation of the maximum eigenvalue statistic. See Enders (2004) for more details.

⁹The Granger Causality test results are not produced here due to space constraint but can be made available upon request.

¹⁰According to Abelson et al. (2005), ‘In an open economy, the exchange rate could influence house prices i.e., a low exchange rate increases the attractiveness of housing assets to foreigners. As a result, the increasing housing investment on housing will lead to rising housing prices in an economy’.

¹¹The causality test results show that there may be a bi-directional causality relationship among two variables, but that could be due to a third common factor with which two variables are related without having true causality relations among the tested variables. Similarly, test result may show that there does not exist any causality relationship, but there could be causality between two variables once the intermediate link between the two variables is established through other variables in a multivariate framework. Therefore, it necessitates a multivariate modelling in order to discover the true direct and indirect relationship among the variables.

¹²It can be emphasized that the current housing price variance (56%) is accounted for by its own house prices in India. This implies housing price is a sufficient statistic for the homebuyers; thereby, the past rise in house prices leads to the present rise in demand for housing with an expectation of a future rise in house prices, and this in turn results in a current rise in house prices. In contrast, for the house builders, the house price may not be a sufficient statistic due to the increasing costs of production, which limits the housing unit production.

¹³The housing sector in developing economies is also not greatly affected by business cycles.

¹⁴Of these two reasons, as cited in the above, the second reason seems to be the sufficient factor in the case of speculative investors in making their investment decisions. They basically prefer to invest in the stock market because it involves high windfall gains backed by a high risk in the short-run, while investing in housing requires a huge lump sum of money and it is long term in nature.