Ben Bernanke has a very good post on the topic of using monetary policy to stabilize asset markets, with the ultimate goal being greater financial stability. I have a post over at MoneyIllusion that comments on his post, but here I'd like to discuss all the hurdles that someone would have to overcome in order to persuade me that it makes sense to add a "third mandate" to monetary policy:
1. Many distinguished economists, including Nobel Prize winners like Eugene Fama, do not believe that government bureaucrats can predict asset price movements more accurately than the market consensus. And yet being able to do so is a necessary condition for being able to effectively smooth asset prices.
2. Unfortunately, while market inefficiency is a necessary condition for a third mandate, it's far from a sufficient condition. If we start with a policy regime aimed at macroeconomic stability (such as inflation targeting, NGDP targeting, or the Taylor Rule) and then add a third mandate, then by assumption you are adding increased macro instability, at least in the short run. (If macro instability did not increase then you'd still (de facto) be focused on the traditional goals of monetary policy.) And macro instability has big costs. Thus you have to have confidence that you could pop bubbles in real time, and that the benefits of popping them would exceed the costs in terms of increased macro instability.
3. But point #2 understates the problem, because macro instability is itself a cause of asset price instability and financial instability. While the beginnings of the recent financial crisis preceded the Great Recession, the crisis got much worse in the US as a result of the recession itself. This is even more true in Europe, where the sovereign debt crisis would likely not have occurred in most countries without the severe drop in eurozone NGDP growth. One direct effect of a third mandate is more macro instability, and one indirect effect of more macro instability is more financial instability.
4. The policy has been tried. During the 1920s New York Fed Governor Strong opposed adding a third mandate, insisting the Fed should focus on stabilizing prices and output growth. After he died in mid-1928, new leadership took over at the Fed and they began raising interest rates with the goal of deflating the stock market "bubble." (I use scare quotes because modern research suggests that stock prices may not have been overvalued in 1929.) At first stocks kept rising, and only crashed when interest rates were raised high enough to drive the economy into a deep depression. This suggests that modest interest rate increases will not significantly impact asset price bubbles, and large increases can have catastrophic consequences for the macroeconomy. And you can't really argue that the Fed should have moved sooner than mid-1928, because there is almost no evidence that stocks were "overvalued" prior to that date. Indeed sophisticated finance theory suggests that stocks have been vastly undervalued for most of the 20th century, producing long run returns above what is justified by the modest risk associated with holding a market portfolio.
To summarize, adding a third mandate (using monetary policy to achieve financial stability) is only justified if:
1. Markets are much less efficient that bureaucrats.
2. Monetary policymakers can pop bubbles so adeptly that the very real direct costs are less than the assumed benefits.
3. We can be sure that bubble popping would never again lead to a macroeconomic disaster like 1929-33.
Those are three very big hurdles. Does that mean do nothing? No, I favor changes in financial market regulation to address the moral hazard problem in the system. What Bernanke calls "the right tool for the job."