Money supply is one of various monetary variables that are influenced by monetary policy. Thus many academic and professional observers hypothesize a close relationship between stock process and the money supply. Sprinkel(1967) use chart analysis and regression method to determine how does monetary policy to affect the stock by empirical fashion. He found that during the period 1918-1963, changes in money supply lead bull market for 2 months. Most of the later scholars also study the issue by using regression analysis. Keran(1976) investigated relationship between the return of stock and changes in money supply during the first quarter in1956 to second quarter in 1970, he found that money supply changes prior to S & P500 index by about two quarters; Homa & Jaffee(1971), and Hamburger & Kochin(1972) also used the regression method and obtained similar conclusion With Keran. These empirical results show that changes in money supply can cause changes in stock market prices, using money supply data can predict future stock price changes. These findings conflict with the efficient market hypothesis, because the efficient market hypothesis is that all known information is already reflected in current share price. However, these studies are basically use monetary stock as a measure of monetary policy within a certain period, which breed serious problems. Empirical studies designed to measure the impact of monetary policy are inconclusive. The problem is that money, prices, and interest rates, as well as Federal Reserve policy, are all endogenous. When the velocity of money change, the money supply will be subject to change .As a result, empirical tests of the relations among these variables are ambiguous without a structural model. Unfortunately, there is not yet a consensus on the appropriate structural model. Not accounting for this endogeneity may introduce a significant bias in empirical estimations of the reaction of equity returns to monetary policy
The previous research is based on the relationship between the monetary stock and stock price. On the other hand, some later studies focus on event studies method. The mainly difference between these two methods are that the event studies method examined how did the monetary policy announcement cause immediate reaction of the stock market. In this respect, the observation interval can be largely reduced. The choice of the data began to daily data and weekly data instead of monthly data or quarterly data in the past, so the announcement of monetary policy relative to the daily stock reaction is exogenous. Tests based on the shorter observation interval are more useful. There is a unique opportunity to study the reaction of asset prices to money supply announcements presents
The following part will review the theories designed to explain the relation between money supply announcements and asset prices. The theories as signs make expectations about the future supply of and demand for money, and the relation between these expectations, real activity, inflation forecasts by people, and the ex ante real rate ,hence the stock price. Basically, there are 3 hypotheses made regarding to this relationship.
Money Supply Announcements and Asset Prices: The Theory
The Expected Inflation Hypothesis
The most obvious link between money announcements and short-term interest rates is through expected inflation. (Eugene Fama, 1975) The hypothesis is that announcement of an unanticipated change in the money supply leads to expectations of higher inflation and thus to an increase in short-term interest rates, while announcement of an unanticipated drop in money has the reverse effect. Gordon’s (1962) growth model assesses the behavior of the price earnings and dividend price ratios. This model says that the current price of a stock, depends on three factors: 1) the dividend the stock paid in the previous period, , the current period expectation of the future growth rate in dividends, , and the current period discount rate relevant for stocks in the firm’s risk category, , as related in equation (1).
Therefore, the stock price is determined by two basic factors: expected dividend or the future cash flow and the expected rate of return or stock discount rate required by the investor. Gordon growth model models illustrate the price of the stock was negatively related to discount rate and positively related to the future dividends or cash flow. One requirement of the expected inflation is that prices respond quickly to changes in money. Rational investor would alter their short-run inflation forecasts on the basis of a money supply announcement. As a result, rates of return on short-term assets such as stock and Treasury bond would be affected. On the other hand, the long-term rate depends on how investors comprehend the money supply announcements. If their expectation regarding future money growth and future inflation will be permanently modified, then long rates will change as much as short rates. Changes in expected inflation would have little impact on stock prices because anticipated corporate cash flows, and the rate at which those cash flows are discounted, would be adjusted by offsetting amounts. As a result, the predicted response of stock prices is not so straightforward and ambiguous. In addition, this prediction breaks down if there are taxes (Martin Feldstein, 1980), or if market imperfections induce firms to contract in nominal terms (Kenneth French, Richard Ruback, and G. William Schwert, 1983). James Tobin (1958) comment that people's risk aversion and their perception of the riskiness of competing assets can determine the money demand. Aggregate risk aversion has risen if the money stock reveals an unexpected change and then substituted assets such as stock and bond are now assumed to be riskier. In this respect, the investors’ required real returns will rise and asset prices to fall. If the information contained in money supply announcements causes risk premiums to increase, it is clear that stock prices will fall. Empirical studies by Charles Nelson (1976), Fama and Schwert (1977), and Schwert (1981), among others, show that stock returns are negatively correlated with both expected and unexpected inflation. However, Robert Geske and Richard Roll (1983) present evidence which indicates that the negative correlation arises because falling stock prices signal higher inflation, not because higher expected inflation leads to a drop in stock prices. To be conservative, therefore, the relation between stock prices and expected inflation is judged to be ambiguous. To summarize, the expected inflation hypothesis predicts that (i) short-term interest rates will rise; long- term rates also will rise, but the increase will be less than for short-term rates unless the change in expected inflation is permanent; (ii) stock prices may rise or move down depending on the role of taxes, nominal contracting by firms and other markets imperfections. In addition, if the aggregate risk aversion had risen, the stock price will fall
B. The Keynesian Hypothesis
The Keynesian liquidity preference model can be written as M/P=f(Y,i),it is a well-known equation. A liquidity effect exists because the price level (p) is sluggish and does not respond instantaneously to monetary shocks. For this reason, real money balances rise as a result of a sudden jump in the supply of nominal money balance. The fluctuation of asset prices can be explained through the expected real rate. Unanticipated jump in money supply will affect the ex ante real rate. The interest rate must fall and offset the incremental money demand. Since expected inflation will rise, the decline in the nominal rate must be due to a drop in the ex ante real rate. In addition, excess money supply causes higher demand for bonds resulting from lower interest. Stock price will increase. However, this liquidity effect is a short-run phenomenon. The real story behind it is that following unexpected increase in money supply central bank may tighten monetary policy by increasing the interest rate of via the liquidity effect. They pursue aggressively for funds and drive up the current interest rate. Urich and Wachtel (1981) refer to this scenario as the policy anticipation effect. Monetary restraint produced an increase in the real rate and leads to lower stock prices for two reasons. First, the discount rate rises to reflect the higher required return by investor. Secondly, expected corporate cash flow will decline because the cost of the interest rate may higher than the return of the investment. All the companies bear the high borrowing rate hence the project is rejected by negative NPV.
C. The Real Activity Hypothesis
Future Output determines the future demand for money, however, no one know what other‘s expectation for money demand, when the money supply was announced, this information provides estimates of the expected future output will rise. The real rate must drive up to clear the market for consumption and investment. Information about the increased future output causes agents to modify upwards their expectation of future real activity. As a result, this information should cause stock prices to go up. Though the discount rate will rise to reflect the higher ex ante real rate, the increase is more than offset by the growth of expected corporate cash flow. This must be the case because the reason for the higher real rate is an expected increase in output. In an empirical study (1981), furthermore, Fama finds a highly significant relation between rates of return on common stock and innovations in real activity. To conclude, the real activity hypothesis predicts that an unexpected increase in the money supply will increase the stock price
In conclusion, Bradford Cornell (1983) uses two different tests of the hypotheses. The first is based on how do announcements of MI and the monetary base influence the response of the prices of short-term bills, long-term bonds, common stock, and foreign exchange. The empirical evidence reveals that none of the above hypotheses explains the reaction of all four assets. The second test, based on the relation between announced innovations in money and expected money, is consistent with the Keynesian hypothesis, but the results can be more readily explained by random variation in float and Treasury deposits.
