If we want to be honest with ourselves and the people who bother to pay attention to what we say, we must acknowledge when we were wrong. I want to do that as clearly as possible. I repeatedly argued against the Fed’s path of rapid rate hikes. I was concerned that the rapid pace of rate hikes would lead to a sharp jump in unemployment.
Ostensibly, the Fed was looking to weaken the labor market (raise unemployment) as a way to reduce the pace of wage growth, and in that way slow inflation. I thought that inflation was likely to come down even without a big jump in unemployment, as the supply chain problems associated with the pandemic were resolved.
It seems that the Fed’s view of inflation was incorrect. The rate of inflation has fallen back nearly to the Fed’s 2.0 percent target, even as the unemployment rate remains below 4.0 percent.
However, I was very much mistaken on the impact of the Fed’s rate hikes. The unemployment rate today is 3.8 percent, only slightly higher than the 3.6 percent rate when it started raising rates last March. Clearly the Fed’s rate hikes did not have the disastrous impact on unemployment I feared.
Why higher rates didn’t raise unemployment
It is possible to identify reasons why the rate hikes did not have as much impact as I and others expected. Usually rate hikes have their largest impact on housing construction.
While the rise in rates did sharply reduce the number of housing starts, from around 1.8 million at an annual rate last March to a bit over 1.3 million in recent months, it did not reduce the number of homes under construction. There were just over 1.6 million under construction when the Fed began raising rates. In recent months the number has been over 1.8 million. In keeping with this increase in homes under construction, employment in residential construction has actually risen since the Fed started raising rates.
The explanation for this seeming paradox is that there was a huge backlog of houses in the pipeline as a result of pandemic supply chain problems. This backlog will eventually be whittled down, but to date, the Fed’s rate hikes have not had the impact on residential construction that would ordinarily be expected.
The second area where Fed rate hikes generally have a large impact is on the trade deficit. This works through a rise in the value of the dollar. The dollar is supposed to rise in value relative to foreign currencies, as people buy dollars in order to take advantage of the high rates here.
This route hasn’t had the usual impact either. The main reason was that the dollar had already risen considerably against other major currencies before the Fed started raising rates. The dollar rose by roughly 10 percent against the euro and a comparable amount against the yen between the start of 2021 and the first Fed rate hike in 2022.
It has risen further against both currencies in the last year and a half, but the trade deficit has nonetheless fallen. It stood at 4.4 percent of GDP in the first quarter of 2022, it was down to 3.0 percent of GDP in the second quarter of this year.
The rise in the dollar surely had some effect in pushing the deficit higher, but this was likely swamped by the effect of consumers shifting away from buying goods following the end of the pandemic. At the height of the pandemic people were unwilling or unable to go to movies, concerts, or travel. As a result, when they spent money it was overwhelmingly on goods consumption, things like cars and TVs. A large share of these goods were imported.
As the impact of the pandemic waned, people shifted back towards buying services and spent a smaller share of their income on goods. The result has been a drop in the trade deficit.
Another area where we expect higher interest rates to have a large effect is on investment in non-residential structures. This category of investment tends to be more interest sensitive than shorter-lived assets, like equipment and software.
Here also the effect of the pandemic lessened the impact. Investment in structures had already fallen sharply, dropping from 3.2 percent of GDP in the fourth quarter of 2019 to 2.6 percent of GDP in the fourth quarter of 2021, a drop of close to 20 percent. Construction of office buildings and retail space fell through the floor as a result of the pandemic, as there was enormous over-supply in both areas.
This meant that as the Fed began raising rates in the spring of 2022 there was not much room for these areas to drop further. In addition, the Biden administration’s polices, notably the CHIPS Act and Inflation Reduction Act, spurred construction of factories producing semi-conductors, batteries, solar panels, and other items needed for a green transition.
These incentives swamped any negative impact from higher interest rates. Construction of factories was 65.9 percent higher in August of 2023 than in August of 2022. Structure investment now stands at 3.1 percent of GDP, almost back to its pre-pandemic share.
The peculiar situation created by the pandemic meant that higher interest rates could not have their normal effect in slowing growth and weakening the labor market. This is why the labor market has remained solid in spite of the sharpest set of rate hikes in more than forty years.
Negative effects of higher interest rates
Even though rate hikes have not produced the slowing that we would ordinarily expect, they still did have an effect on the economy. The big jump in interest rates essentially shut down mortgage refinancing. The low mortgage rates of the pandemic allowed roughly 14 million homeowners to refinance their mortgage between 2020 and 2022.
According to research from NY Federal Reserve Bank, five million of these borrowers took out a total of $430 billion in equity, which they used to support their consumption or invest in other assets. The other nine million borrowers saved an average of $2,500 a year on interest payments by refinancing at lower rates. Higher interest rates put an end to the refinancing boom, with refinancing down more than 90 percent from its pandemic peak.
Higher mortgage rates also put a squeeze on homebuying. Sales of existing homes are down by almost a third, more than 2 million at an annual rate, from their levels in February of 2022, before the Fed began raising rates. The drop in existing home sales does have some impact on the economy. When people buy a home it generates fees and commissions for realtors, mortgage issuers, and various other actors involved in a home sale. People often tend to buy things like refrigerators and dishwashers when they buy a house and possibly remodel or paint their new home. For this reason, the drop in existing home sales does slow growth, even if the impact is much smaller than would be the case with a comparable drop in the sale of new homes.
While this fall in spending is picked up in GDP, there is another aspect to the drop in home sales that is not picked up in our GDP measures. The reduction in home sales is mostly a story where people would like to sell their current home, and move to a new one, but are reluctant to do so because it would mean giving up a mortgage with a very low interest rate, and taking out a mortgage on a new home with a much higher interest rate. As a result, they put off moving to a home that might better fit their needs.
As was pointed out to me by Adam Ozimek, this is a real cost to higher interest rates that is not picked up in GDP. People who would otherwise be in a different home are unambiguously worse off as a result of high current mortgage rates. (A huge gain that is not picked up in GDP is the increase in the number of people working from home, who are saving thousands of dollars a year on commuting costs and hundreds of hours of commuting time. The number of people working from home has increased by more than 11 million since the pandemic.)
Another cost is that many smaller firms and start-ups are having more difficulty getting access to capital. This may not be a big deal in terms of current investment, but if many of these firms are more innovative than larger incumbent firms, we may be paying a price down the road in the form of less innovation and productivity growth.
And, we know that higher rates have produced stress in the financial system. The wave of bank failures that started with the collapse of the Silicon Valley Bank was a predictable outcome from the sort of sharp rise in interest rates we have seen over the last year and a half. While banks should have hedged themselves from interest rate risk, it is impossible to do so completely, and many financial institutions will be facing serious stress as long as rates are high.
Lowering rates and getting back to normal
For these reasons, it would be desirable to see the Fed start to turn the corner on interest rates. We don’t really need lower rates to boost the economy just now, we look to be on a healthy growth path for the foreseeable future. But we would nonetheless see substantial benefits from a decline in interest rates.
The lesson from the limited impact of the sharp rise in rates on growth should also apply in reverse. Lower rates will clearly have a positive impact on residential and non-residential construction, as well as the trade deficit, but it the impact is not likely to be as large as previously believed. Just as was the case with the rise in rates, other factors are likely to be more important in determining demand in these areas.
To be clear, there is no reason for the Fed to do a sharp reversal on rates. The economy is not in desperate need of stimulus. But we should be looking to get back to something resembling normal following the steep pandemic recession and the sharp recovery. This means edging down to the sort of interest rate curve we saw before the pandemic. A statement of this intention by the Fed, along with a modest rate cut, would be a huge step in this direction.