Another great post. Of course, if we don't know what determines inflation, we have the same problem with exchange rates. If PPP holds - which you assume in writing an open econ Fisher relationship, then P=EP*, in %change, inflation = %E + inflation*. Real int parity says r=r* --> i+inf = i*+inf* (inflation here is Expected) so we get i - i* = -(inf - inf*)=-%E. As you state. So, in many ways this column is just a restatement of your last column. That being said, I think your column is 100% right. All the failures in the NK and other models translate directly into open economy versions. At a deeper level, I think one missing piece - at least for me - is the distinction between what determines steady state inflation versus what determine deviations of inflation around a constant steady state. I think it's a point often overlooked because we usually assume steady state inflation is zero. But break open an NK model and nothing determined s.s. inflation, just that "it is the target". Push harder and they'll say, okay, assume an ad hoc real money demand equation and growth rate of M equals growth rate of P in steady state. ... Hmm. ... Anyway....Please keep pushing this line Prof Cochrane. I really think you are going in the right direction. Thanks.
Think of a (real) fed rate change as a change in the short term level of subsidy for holding dollars at the federal reserve. Higher subsidy, more short term demand. More money at the fed, less money supply. Less money supply, certain activities like real estate construction that require loans decrease. And then a decrease in real economic activity that decreases inflation. More money is printed, but because it subsidizes parking money at the fed with fractional reserve banking it can actually dramatically reduce the money supply. Reduced money supply increases demand for the currency and thus improves exchange rates in the short run. In the long run though you printed money so in the long run inflation goes up and the currency depreciates.
The key is that unlike when the fiscal side of the house prints more money, the fed in a rate increase uses its printed money to temporarily pay people to decrease the money supply and eventually slow the economy. So in the short run, foreign assets are used to buy the currency to farm the subsidy, which is paid not just by the fed, but by productive members of society with now higher capital costs, but in the medium term the trade must unwind because the economy starts to sour making investments riskier, and in the long run more money leads to currency depreciation.
The key insight is to ask what is being subsidized over what time horizon and to recognize that printing money can temporarily reduce the money supply if it’s used to pay banks to not make loans. In the short term this can be partially offset by foreign exchange chasing higher rates, but in the long run more of a currency makes it less valuable.
On the face of it the yen has been on the solid line since the BOJ hiked from 0-0.1% to around 0.25% on July 31, appreciating from 160 to 140 against the dollar. There was no fiscal tightening. The BOJ rate hike was accompanied by an announcement that it would scale down its bond buying but why should the ownership of the bonds matter for the yen (or anything else)?
Oops. The question looks like we are starting by “Reasoning from a price change” [™ Scott Sumner.]
Why are interest rates rising? Do we mean that starting from macroeconomic dynamic equilibrium (relative prices are not changing), a, say, tax cut for the highest income people, causes a deficit. In order to hold inflation at its target (not an interest rate “target;” interest rates are policy instruments), the Fed has to raise interest rates to suppress some domestic investment and attract some foreign investment to accommodate the increased consumption by the tax-favored folks? In this circumstance, yes, the trade deficit will expand. The inflow of capital will depress the exchange rate and incomes of exporters and “China shocked” import competing producers and Owen Cass can have a hissy about neoliberalism.
Another great post. Of course, if we don't know what determines inflation, we have the same problem with exchange rates. If PPP holds - which you assume in writing an open econ Fisher relationship, then P=EP*, in %change, inflation = %E + inflation*. Real int parity says r=r* --> i+inf = i*+inf* (inflation here is Expected) so we get i - i* = -(inf - inf*)=-%E. As you state. So, in many ways this column is just a restatement of your last column. That being said, I think your column is 100% right. All the failures in the NK and other models translate directly into open economy versions. At a deeper level, I think one missing piece - at least for me - is the distinction between what determines steady state inflation versus what determine deviations of inflation around a constant steady state. I think it's a point often overlooked because we usually assume steady state inflation is zero. But break open an NK model and nothing determined s.s. inflation, just that "it is the target". Push harder and they'll say, okay, assume an ad hoc real money demand equation and growth rate of M equals growth rate of P in steady state. ... Hmm. ... Anyway....Please keep pushing this line Prof Cochrane. I really think you are going in the right direction. Thanks.
I'm really not that interested in working through the implications of a non-FIAT Fed.
Think of a (real) fed rate change as a change in the short term level of subsidy for holding dollars at the federal reserve. Higher subsidy, more short term demand. More money at the fed, less money supply. Less money supply, certain activities like real estate construction that require loans decrease. And then a decrease in real economic activity that decreases inflation. More money is printed, but because it subsidizes parking money at the fed with fractional reserve banking it can actually dramatically reduce the money supply. Reduced money supply increases demand for the currency and thus improves exchange rates in the short run. In the long run though you printed money so in the long run inflation goes up and the currency depreciates.
The key is that unlike when the fiscal side of the house prints more money, the fed in a rate increase uses its printed money to temporarily pay people to decrease the money supply and eventually slow the economy. So in the short run, foreign assets are used to buy the currency to farm the subsidy, which is paid not just by the fed, but by productive members of society with now higher capital costs, but in the medium term the trade must unwind because the economy starts to sour making investments riskier, and in the long run more money leads to currency depreciation.
The key insight is to ask what is being subsidized over what time horizon and to recognize that printing money can temporarily reduce the money supply if it’s used to pay banks to not make loans. In the short term this can be partially offset by foreign exchange chasing higher rates, but in the long run more of a currency makes it less valuable.
On the face of it the yen has been on the solid line since the BOJ hiked from 0-0.1% to around 0.25% on July 31, appreciating from 160 to 140 against the dollar. There was no fiscal tightening. The BOJ rate hike was accompanied by an announcement that it would scale down its bond buying but why should the ownership of the bonds matter for the yen (or anything else)?
Oops. The question looks like we are starting by “Reasoning from a price change” [™ Scott Sumner.]
Why are interest rates rising? Do we mean that starting from macroeconomic dynamic equilibrium (relative prices are not changing), a, say, tax cut for the highest income people, causes a deficit. In order to hold inflation at its target (not an interest rate “target;” interest rates are policy instruments), the Fed has to raise interest rates to suppress some domestic investment and attract some foreign investment to accommodate the increased consumption by the tax-favored folks? In this circumstance, yes, the trade deficit will expand. The inflow of capital will depress the exchange rate and incomes of exporters and “China shocked” import competing producers and Owen Cass can have a hissy about neoliberalism.
The standard model looks pretty good to me.
Scott Sumner responds here:
https://www.econlib.org/cochrane-on-interest-rates-and-exchange-rates/