Choosing the right tools
By Scott Sumner
Here’s the always brilliant Steven Landsburg discussing the importance of theoretical tools:
I sometimes hear economists defend the unrealism of their models thusly: “Economics is an infant science. Today our models are unrealistic; a decade from now, they’ll be a little so, in a decade from then little less. Eventually we’ll have realistic models that make accurate predictions.”
That, I think, is pure poppycock. Our predictions are not, and never will be, based on models; they’re based on informal reasoning. We study models because they own our reasoning skills. We can figure out what happens in these models and thereby develop a good intuitive feeling for what sorts of reasoning are likely to be productive.
Over at TheMoneyIllusion I also provided this quotation from his book The Big Questions, and listed 47 tools that I sometimes use for money/macro problems. I won’t repeat the very long list over here, but you can check out the link if interested. Instead I’d like to show how macroeconomists go about applying these tools. Or at least how I do; I can’t speak for others.
Let’s take the example of Federal Reserve “forward guidance.” How do we think about this policy? I will bold each tool I use from my MoneyIllusion list.
Let’s begin with an assumption that next week the Fed issues a surprise announcement; the fed fund target will not be raised at all during 2015. How should we think about that announcement?
1. The motivation for using forward guidance as a policy tool is the zero interest rate bound, which prevents the Fed from cutting short term rates today.
2. Because of the identification problem, a promise to hold interest rates low for an extended period might be expansionary of contractionary, depending on the situation. In order to determine whether the Fed was signaling faster NGDP growth as everyone but the NeoFisherians assumes, and whether the policy had credibility, we would employ the efficient markets hypothesis and rational expectations, and look at the response of TIPS spreads and stock prices. If they rose, the policy was probably intended to be expansionary–a signal of higher future NGDP growth.
3. Now we have to figure out how the Fed would achieve this policy. Most likely, they would rely on the liquidity effect. Today markets expect the Fed to begin raising rates in the summer of 2015. In order to hold rates down to zero until the end of 2015, they would have to rely on the liquidity effect of increases in the monetary base. Thus this forward guidance should be interpreted as a promise to increase the monetary base in late 2015 more rapidly than expected, at least for any given level of money demand. (Here things are quite complicated, as the announcement might also shift money demand next year. And they might use IOR.)
4. Next let’s think about the impact of a larger monetary base next year. According to the hot potato effect people would try to get rid of these excess cash balances, driving down the value of money (which might be measured in terms of 1/P, 1/NGDP. or 1/ forex prices.) As the value of money falls, the price level, NGDP and forex prices rise.
5. The faster inflation might tend to raise longer-term interest rates, via the Fisher effect. Exchange rates will fall in the long run, due to purchasing power parity (recall the price level rises due to the hot potato effect, AKA quantity theory of money.) In the short run, exchange rates will fall even more sharply than in the long run, as explained by Dornbusch’s overshooting model. This is because the one year bond yield will fall due to the liquidity effect (by assumption), and the interest parity theory suggests that lower nominal interest rates imply a stronger expected appreciation in the domestic currency. Since the long run effect is currency depreciation, in the short run the currency must fall even more sharply, so that it can be expected to gradually appreciate due to interest parity. (Confused?–don’t worry, it’s gets much easy now.)
6. Now let’s think about aggregate demand (AKA NGDP.) The increase in expected future NGDP growth will raise the Wicksellian equilibrium interest rate, making monetary policy more expansionary. This will tend to boost current AD and current NGDP. Because of sticky wages, rising NGDP will reduce the ratio of wages/NGDP. (Wages are sticky due to money illusion and transactions costs.) This means firms will move down and to the right along the downward-sloping labor demand schedule, leading to higher levels of employment.
7. Due to Okun’s Law, higher employment will tend to raise real output. In addition, medium and longer-term interest rates may rise due to the income effect and the expectations hypothesis of the term structure of interest rates. Investment will probably rise faster than consumption due to consumption smoothing.
8. Unfortunately, the Greeks are not able to engage in forward guidance, due to policy impotence under fixed exchange rates. However, the natural rate hypothesis predicts that money is neutral in the long run, and hence that Greek unemployment should return to their natural rate once wages and prices adjust. But this adjustment will not resolve their debt crisis, which reflects the interaction of nominal debt contracts and falling NGDP.
The trick is to think deeply about which tool applies to which situation. I find it useful to always begin with an equilibrium (flexible price) analysis to get a sense of the long run equilibrium. Then work backwards from the long run by adding sticky wages and prices, and or nominal debt contracts and unindexed capital taxation regimes. Always clearly distinguish between real and nominal effects. Yes, many variables such as GDP and exchange rates feature high levels of correlation between real and nominal in the short run, in some countries (but not Zimbabwe). But these variables are causally unrelated in the long run, and hence it’s dangerous to blindly assume you can use these terms synonymously.
Never think about “money flowing” into or out of asset markets, or “selling waves” hitting Wall Street. Asset purchases equal asset sales. Think in terms of equilibrium asset prices, reflecting both local and macro factors.