Twin Booms: The Lead–Lag Relation Between Credit and Housing Booms

Abstract

In this study, we examine the causal relation between credit (proxied by credit-to-GDP ratio) and house markets (proxied by house price index) using data of using thirty-six countries for the period 1996–2012. We find a bidirectional causal relation between the two markets using the whole sample. Then, we find that during the non–twin boom period, the results are the same as those using the whole sample. During the twin boom periods, the two markets are not linked using the boom definition of deviation from the trend, and the housing market leads the credit market using the boom definition of the growth rate exceeding a certain 15 percent.

KEY WORDS:

• credit booms
• housing booms
• panel Granger causality

Notes

1. Eichengreen and Arteta (2002), Demirgüç-Kunt and Detragiache (1998, 2005), Hume and Sentance (2009), Jordà, Schularick, and Taylor (2011, 2013), Kaminsky and Reinhart (1999), Mendoza and Terrones (2012).

2. Glaeser, Gottlieb, and Gyourko (2010) indicate that the house price boom experienced across many Organisation for Economic Co-operation and Development (OECD) countries appears to have an increasing correlation across countries.

3. Levine (2005) argues that in the growth literature, financial development is proxied by private credit as a share of GDP, while a credit boom is identified by an abnormally high growth rate in that same variable.

4. See Barajas, Ariccia, and Andrei (2007), Dell’Ariccia, Igan, and Laeven (2012), Gourinchas et al. (2001), Kaminsky and Reinhart (1999), and Williams (2012).

5. See Shen et al. (2011) for a survey.

6. Dell’Ariccia, Igan, and Laeven (2012) define the start of the boom as the earliest year in which either the credit-to-GDP ratio exceeds its trend by more than three-fourths of its historical standard deviation while its annual growth rate exceeds 5 percent or the annual growth rate exceeds 10 percent. A boom ends as soon as either the growth of the credit-to-GDP ratio turns negative or the credit-to-GDP ratio falls within three-fourths of one standard deviation from its trend and its annual growth rate is lower than 20 percent.

7. Agnello and Schuknecht (2011) use HP with a smoothing parameter of 10,000. We do not use cubic trend and HP method because of the small sample size for housing price of each country.

8. The thirty-two countries are Argentina, Australia, Austria, Belgium, Brazil, Canada, Chile, China, Denmark, Finland, France, Germany, Greece, Hong Kong, Iceland, Indonesia, Ireland, Italy, Japan, Korea, Malaysia, New Zealand, Philippines, Portugal, Singapore, Spain, Sweden, Switzerland, Taiwan, Thailand, United Kingdom, and United States.