Thanks for this! I oversee the Macroeconomic Stabilization grant portfolio at Open Phil. We’ve been reevaluating this issue area in light of the current macroeconomic conditions and policy landscape, and we’re planning to write more about our own perspective on this in the future—so I won’t reply line by line here. But we’re always eager for substantive external critique, so I wanted to flag that we’d seen this and appreciate you sharing!
One clarifying question: are you suggesting that we could reduce inflation risks without running higher unemployment in expectation? From my perspective, a dovish macroeconomic policy framework has both costs and benefits. We’re generally going to face too-high inflation at one part of the business cycle and too-high unemployment at another part of the cycle. A philanthropist could push for a more dovish framework in order to minimize the UE overshoot at the expense of risking more inflation overshoot. I see those two risks as trading off against each other—curious if you agree.
Thanks for engaging with this! I agree with all your points, but with some important qualifications.
Yes, in normal times, we can’t reduce inflation risks without risking losing any jobs and that there’s usually a trade-off. And so, as a rule, if we push for more dovish policy to lose fewer jobs, then we risk overshooting on inflation. Yes, there’s usually too-high inflation at one part of the cycle and too-high unemployment at another—it’s never optimal. But it depends on how ‘too-high’. We live in extraordinary times (even antebellum).
The US hasn’t had such high inflation and such low unemployment since the 60s, and surprisingly now, the flat Phillips curve need not imply that reducing inflation will cost many jobs at all [see recent paper by Fed officials]. Even Krugman now says that the job market is running unsustainably hot and says that this time reducing inflation won’t risk high unemployment (though the landing won’t be soft) and also ‘Cooling that market off will probably require accepting an uptick in the unemployment rate, although not a full-on recession.’ [src] The labor market is just very tight right now [also see this empirical paper]. And so, I don’t think that we always want to ‘minimize unemployment’. The jobs we’re adding are not probably not good jobs.
And yes, as you say, dovish policy has costs and benefits. And indeed, it created a lot of benefits up until a point and then it’s about the marginal cost-effectiveness. Because of diminishing returns, the high costs on the ’21 margin outweighed the benefits. So in essence, we way overshoot. As I write above:
“‘Wise owls were dovish in the 2010s and are hawkish today. Don’t be a permahawk or a permadove.’
To paraphrase Furman: Both the position that loose policy created lots of jobs, income, and growth, but not much inflation, nor the converse position that loose policy caused lots of inflation but little growth are wrong. Rather, we got more and more inflation and fewer and fewer jobs for each dollar we added. The first $1T of stimulus led to a great recovery, but the last $0.5T of stimulus caused a lot of inflation but only few jobs. In other words, the marginal cost-effectiveness of looser policy was bad and the stimulus should have been smaller—maybe $1 trillion instead of 2.”
Indeed, empirically we now see that the US labor market is now very tight, i.e., close to full employment and hasn’t added many jobs recently. This is another sign that we’ve hit diminishing returns to expansionary (looser) policy.”
And maybe this could have been avoided. An econometrics paper from May ’21, estimated a 32% chance of >6% inflation in ’22 and warned that the discussion was based on casual observation rather than econometrics [src].
And so it’s not a matter of dovish vs. hawkish, but a matter of degree. We were too dovish and overshoot.
Thanks for this! I oversee the Macroeconomic Stabilization grant portfolio at Open Phil. We’ve been reevaluating this issue area in light of the current macroeconomic conditions and policy landscape, and we’re planning to write more about our own perspective on this in the future—so I won’t reply line by line here. But we’re always eager for substantive external critique, so I wanted to flag that we’d seen this and appreciate you sharing!
One clarifying question: are you suggesting that we could reduce inflation risks without running higher unemployment in expectation? From my perspective, a dovish macroeconomic policy framework has both costs and benefits. We’re generally going to face too-high inflation at one part of the business cycle and too-high unemployment at another part of the cycle. A philanthropist could push for a more dovish framework in order to minimize the UE overshoot at the expense of risking more inflation overshoot. I see those two risks as trading off against each other—curious if you agree.
Thanks for engaging with this! I agree with all your points, but with some important qualifications.
Yes, in normal times, we can’t reduce inflation risks without risking losing any jobs and that there’s usually a trade-off. And so, as a rule, if we push for more dovish policy to lose fewer jobs, then we risk overshooting on inflation. Yes, there’s usually too-high inflation at one part of the cycle and too-high unemployment at another—it’s never optimal. But it depends on how ‘too-high’. We live in extraordinary times (even antebellum).
The US hasn’t had such high inflation and such low unemployment since the 60s, and surprisingly now, the flat Phillips curve need not imply that reducing inflation will cost many jobs at all [see recent paper by Fed officials]. Even Krugman now says that the job market is running unsustainably hot and says that this time reducing inflation won’t risk high unemployment (though the landing won’t be soft) and also ‘Cooling that market off will probably require accepting an uptick in the unemployment rate, although not a full-on recession.’ [src] The labor market is just very tight right now [also see this empirical paper]. And so, I don’t think that we always want to ‘minimize unemployment’. The jobs we’re adding are not probably not good jobs.
And yes, as you say, dovish policy has costs and benefits. And indeed, it created a lot of benefits up until a point and then it’s about the marginal cost-effectiveness. Because of diminishing returns, the high costs on the ’21 margin outweighed the benefits. So in essence, we way overshoot. As I write above:
And so it’s not a matter of dovish vs. hawkish, but a matter of degree. We were too dovish and overshoot.
Hard it may be to accept, but maybe Manchin was right and we shouldn’t have printed so much money and had such a big stimulus, and we all made fun of the guy on the left here:
but maybe he was onto to something.
Thanks a ton for this clarification! Very helpful.