The first of three mid-year reports from Blockworks Research is an analysis on the headwinds crypto faced in Q2. Read it here for free.
This week, get 50% off a Blockworks Research subscription (granting you full access to exclusive data and insights) with code NL50.
“To try effectively to wipe out hard-core inflation by squeezing the economy is possible but disproportionately costly. It is burning down the house to roast the pig.”
— Robert Solow
When a Washington Post survey asked whether the 1975 Public Affairs Act should be repealed, 43% of respondents agreed it should.
They needn’t have worried: There was no such thing as the Public Affairs Act.
The Post’s poll was a test of George Bishop’s academic work which found that “a substantial number of people will offer opinions on fictitious topics in the context of a survey interview.”
Bishop’s ideas on the “illusion of public opinion” have been a topic in political science since the 1980s.
But they do not seem to have filtered up to the Fed as of yet.
At last month’s FOMC meeting, after carefully guiding the market to expect a raise of 50 bips, the Fed hit the panic button and went 75.
They did so largely, it seems, on the basis of a University of Michigan poll that showed five-year inflation expectations had risen to a multi-decade high of 3.3%.
That number was derived from the total of 300 responses. And it was a preliminary figure, unadjusted for demographics.
Political pollsters are, of course, always tinkering with their survey-based models to make them more predictive, and that is a pretty straightforward process: If 90% of your respondents happen to be registered Libertarians, your unadjusted model might predict Ayn Rand will be the next president of the United States, but I wouldn’t bet too much on it.
How, though, do you adjust a model for inflation expectations? Is the type of person who answers a University of Michigan survey more or less likely to worry about long-term inflation than the average American?
For that matter, how do you get a survey of 300 people to tell you anything at all useful about a country of 330 million people? Even Nate Silver would struggle with that one, I think.
And yet, the Fed thought it was solid enough evidence that a change in the course of monetary policy was urgently needed.
So urgently, in fact, that they couldn’t wait around to see what the University of Michigan’s final number would be.
After adjusting for the fact that the type of person who enjoys responding to surveys tends not to be entirely normal, the university revised its preliminary estimate of 3.3% down to 3.1%.
And this month it fell to 2.8%.
So, to recap: The Fed hit the panic button based on an unadjusted survey of 300 people that returned an anomalous result which one-month later has reverted back to the long-term norm.
Does it matter? Maybe not. We don’t really know.
Which is exactly the problem: We don’t know if inflation expectations matter, we don’t know if Fed policy matters to inflation expectations, and we don't even know how much rate hikes will matter for the economy.
The rush to speed up tightening in response to a single data point implies a level of analytical confidence and precision that should never be assumed in monetary policy — there’s a reason why economics is considered a social science, not a real science.
Fed economist Jeremy Rudd contends that “our understanding of how the economy works — as well as our ability to predict the effects of shocks and policy actions — is in my view no better today than it was in the 1960s.”
Like Bishop’s “illusion of public opinion,” Rudd asserts there’s an “illusion of control” in monetary policy that is “arguably more likely to cause problems than an actual lack of control.”
Powell acknowledged as much in answering a question about unemployment and recession in today’s press conference: “Anyone who is really sure that it's impossible or really sure [what] will happen is probably underestimating the level of uncertainty.”
I was happy to hear that, and the market was even happier to hear he’s aware there are signs the economy is cooling off.
He made clear, however, that the Fed is willing to accept a recession in its quest to fulfill the “stable prices” half of its dual mandate.
And maybe that’s the right thing to do — it may well be best to play it safe with inflation expectations.
But, per Robert Solow, roasting the inflation pig may also risk burning down the economy house.