The strong shift of private savings towards the stock market suggested that the 1990ps boom in consumption was financed by heavily relying on stock market performances. Looking at the U.S., between 1995 and 2000 equity prices rose by about 200% while stock market capitalization as a share of GDP went from 76 to 180. A similar growing pattern characterized the ratio of equity holdings to total financial wealth in house holds portfolios, which steadidy rose from 13% in 1985 to the peak level of 28% just before the early 2000 stock market crash. The Federal Reserve appeared then seriously concerned about the links between financial and real stability, as well as the perils that irrational exuberance might have exerted over consumer and investor confidence and thereby on real activity. In this context, a lively debate started in the economic literature aimed at defining the appropriate response of monetary policy to large swings in stock prices.
The financial accellerator model of Bernanke and Gertler (1999,2001) is probably the most prominent example of a fully fledged DSGE model with supply side linkages between the macroeconomy and the financal market. However, while it captures a quantitatively significant component of aggregate fluctuations due to credit market distortions, this framework remains completely silent on the demand side channel on which the Fed itself expressend some concern. As a matter of fact, there have been very few attempts to model the wealth effects on consumption coming from financial assets in DSGE environments. Notable exceptions are Iacoviello (2005), Iacoviello and Neri (2008) and Monacelli (2008). However, their analysis confines to the role of durables as collaterals and their implications for monetary policy, without any mention of stock market wealth.
We try to fill this gap by presenting a New Keynesian DSGE model where stock price fluctuations distort aggregate consumption via real wealth effects. We consider an economy where financial markets feature a constant turnover between long time traders (holding assets) and newcomers (entering the market with no wealth at all). This heterogeneity in house holds portfolios implies a sharp difference between the laws of motion of individual and the aggregate consumption. More specifically, individual consumption smoothing does not carry over in aggregate terms as the population currently in the market differs from the one that will operate tomorrow. Under this turnover, for a given stock of wealth, expected aggregate consumption is lower than the one that would have occurred in the standard infinite horizon representative agent economy. We show that the wedge between the current and the expected level of aggregate consumption is driven not only by the ex ante real interest rate as in the standard representative agent economy but also by the stock of wealth accumulated today, since the latter is responsible for the difference between the consumption level of long time traders and newcomers.
In this set up, stock market booms (bursts) induce a larger increase (decrease) in current consumption with respect to the standard infinitively lived case. By making the dynamics of aggregate financial wealth relevant for current aggregate consumption as well as for current inflation, this demand side channel intro duces some real effects of stock prices volatility. One of the appealing features of our model is its tractability. As the turnover rate goes to zero, only long time traders operate in the market, and our model resembles a standard infinitively lived representative agent economy.
This demand side linkage between the stock market and the real economy motivates us to study the stabilizing properties of monetary policies that explicitly respond to stock prices.6 In principle, a central bank aiming at price stability might want to increase the nominal interest rate, in response to higher stock prices, since the latter via their positive impact on current demand are eventually inflationary. The objective of this paper is to assess under what conditions this policy which will take the form of an extended Taylor type interest rate rule does not generate aggregate (real) instability that is entirely driven by expectations and completely unrelated to fundamentals, a situation where the Rational Expectations Equilibrium (REE) is said to be indeterminate. Following the approach of Evans and Honkapohja (2001), we also study whether the fundamental REE (often referred to as the Minimum State Variable solution, as in McCallum (1983)), can be eventually attained were agents forming expectations via simple recursive least squares learning because missing sufficient information to make perfectly rational forecasts. This stability under learning (or E stability) analysis is meant to further refine the determinacy results, since, even if determinate, the (fundamental) REE is not necessarily attainable once agent expectations slightly depart from full rationality.
Download
PDF Ebook Learning, Monetary Policy and Asset Prices
