As time passes and the US economy recovers, jitters about the Fed exit are becoming more and more frequent. Figuring out the impact of the exit on markets requires a little knowledge about the response of asset returns to the past policies of the Fed. There may not be a pure symmetrical answer, but at least we can identify a channel.
The Fed's policy has been forward guidance and quantitative easing. The reaction of equity markets to the expansion of the Fed's balance sheet has not been straightforward, as can be seen in the chart below. Interestingly, markets have almost always anticipated the periods of bond purchases. Since 2012, the link has clearly broken down:
The US stock market has been much more sensitive to inflation/reflation expectations - even though the link has broken down in the recent past (the chart shows the link between stocks 3-month return and the change in breakeven inflation spreads over the same period).
Among the common false assertions heard over the last few years, the idea that stocks will rise regardless of the state of the economy thanks to the Fed's intervention is far from true. I have shown several times that 3-month stock returns were highly correlated to 3-month changes in the ISM value. The chart below shows that the S&P 500 returns over 3 months depend highly on the job market (initial claims). It just means that whatever the quantity of assets purchased by the Fed, stocks respond negatively to bad economic news.
The US stock market seems to be much more related to the economic news flow than to the Fed's balance sheet expansion and - maybe temporarily - to inflation expectations.
The impact of the Fed exit on stock prices will depend on the level of stock valuation and the sensitivity to interest rates. The chart below suggests that the interest rate risk for stocks is much higher when valuations are stretched: in the late 1990ies and in 2007, the Fed tightened while stocks' Price Earning Ratios were already close to (or above) levels considered as "rich". This is clearly not the case today.
In addition, the equity risk premium (the difference between earning yields and long bond yields) has been biased by the artificially low level of US 10-year yields (Fed purchases). For this reason I factored into my Equity Risk Premium (ERP) a model-implied fair level of UST 10-year yields. The result is shown in the chart below. I found that the ERP is 1) still slightly above its medium run average, and 2) well above the level implied by the high yield market (see below and here).
There is no doubt that since valuations are not stretched, and the ERP is not too low, the interest rate risk is less significant than what is currently being fueled by the rumor mill. In addition, the separation principle recently adopted by the Fed suggests that it can stop its bond purchase program while maintaining low rates for a longer period.
So long as the US news flow continues to support stocks, and Bernanke continues to communicate the Fed's exit plans coherently, the stock rally should not be "killed" by the Fed.