Credit bubble

A credit bubble is a sustained and accelerating growth of bank loans relative to the growth of Gross Domestic Product (GDP), which finances a boom in both economic activity and in asset prices. The proposition that this growth of credit adds to demand - especially for financial assets - above and beyond that generated from existing incomes contradicts the "loanable funds" vision of lending in which loans are "pure redistribu- tions" which "should have no significant macroeconomic effects" (Bernanke, 2000, p. 24), as lending simply redistributes spending power from lender to borrower without enhancing aggregate demand. However, in the endogenous-money view, lending enables demand to increase in the aggregate, thus financing a growth in economic activity and rising prices on asset markets. Prior to the global economic and financial crisis that erupted in 2008, the dominant view in economics was that the proposition that "credit bubbles" had any macroeconomic significance was a figment of the imaginations of non- economists. The Modigliani-Miller theorem (Modigliani and Miller, 1958) - the relevant subset of the efficient markets hypothesis - argued for the irrelevance of credit to both the valuation of firms (except for the effect of the tax-deductibility of interest payments) and economic performance. The proposition that there could be a "financial accelera- tor" (Bernanke et al., 1996) gave conventional theory an argument as to how credit could impact on economic activity, but this mechanism relied on agency costs owing to asym- metric information and acted through the price of credit rather than its volume.

Since the outbreak of the economic and financial crisis in 2008, this "irrelevance view" has lost favour, and attention has turned to what the empirical record indicates is the effect of greater-than-mean increases in credit. Three authors, namely Jordà, Schularick and Taylor, have led the way here (see Schularick, 2009; Jordà et al., 2011a, 2011b; Schularick and Taylor, 2012), via the construction and analysis of a database of credit and economic cycles for 14 countries over a 139-year period (from 1870 to 2008). They provide an empirical definition of an "excess credit" variable (x) as "the rate of change of aggregate bank credit (domestic bank loans to the nonfinancial sector) relative to GDP, relative to its mean, from previous trough to peak" (Jordà et al., 2011b, p. 5), measured in percentage points per year. Clearly this can range from well below to well above zero (its average value over their sample of recessions was 0.47 per cent of GDP per annum). Their database (www.aeaweb.org/aer/data/april2012/20091267_data.zip) includes 223 recessions, 173 of which they classify as "normal recessions" on the basis of the low value of their x indicator (the average value of x was 0.29), and 50 of which they classify as "financial recessions" (the average value was 0.71). Via a series of econometric tests they conclude that the value of x was the best indicator of the severity of the ensuing recession: the larger its value, the deeper the recession, and the longer its period. Since the value of x prior to a recession accurately indicates the severity of the recession itself, and the value of x invariably plunges during and in the aftermath of the recession, the rise and fall of x can be taken as a manifestation of the expansion and collapse of a credit bubble. The bubble clearly has impacts on both asset and commodity markets, causing a boom in both markets prior to the recession and a bust in both markets during the recession.

They note that their results are consistent with the Fisher (1933) debt deflation hypoth- esis and Minsky's (1963, 1972) financial instability hypothesis, though their empirical research deliberately lacks a theoretical foundation, since a theoretically agnostic position allows them to simply document these "new important facts about the role of credit in the modern business cycle" (Jordà et al., 2011b, p. 38).

Though theoretically agnostic, the fact that "excess credit" (x) is a reliable indicator of both an approaching financial recession and its likely severity challenges the conven- tional view of the role of central banks (that financial crises cannot be predicted ahead of schedule) and that the proper role of central banks is to mop up after crises rather than attempt to prevent them from happening.

Support for the concept of credit bubbles also comes from the work of Biggs, Mayer and Pick (Biggs and Mayer, 2010; Biggs et al., 2010) on the "credit impulse" and Keen (2013, pp. 247-9) on the "credit accelerator". Both terms refer to the change in debt divided by GDP, a measure that is consonant with Schularick's x. These authors propose a causal link between debt acceleration and change in economic activity in both goods and asset markets. This implies that economic booms and rising asset prices in part rely upon accelerating debt, and since continuous positive acceleration is impossible, a credit bubble will be defined by the transitions from positive to negative debt acceleration.

See also:

Asset price inflation; Asymmetric information; Bubble; Debt deflation; Efficient markets theory; Endogenous money; Financial bubble; Financial crisis; Financial instability hypothesis; Money and credit; Minsky, Hyman Philip.