Larry Summers isn’t worried about secular stagnation anymore

But the fundamental drivers are still in place

Reminder: If you haven’t filled out our reader survey yet, please do! It only takes three minutes, and five participants will win Slow Boring merch.

Slow Boring staff is on spring break this week, but we’re excited to share some fantastic content with you while we’re gone. Today’s guest post is from Slow Boring alum, Milan Singh, who shares my somewhat idiosyncratic interest in the long-term trajectory of the debate over the long-term structural determinants of interest rates. Is the world still facing “secular stagnation” or are today’s higher rates here to stay? And why?

---

Back in 1939, Alvin Hansen wrote a paper titled “Economic Progress and Declining Population Growth.” He argued that while population growth led to lower living standards before the industrial revolution, we now live in Adam Smith’s world and a growing population is the fundamental driver of economic growth.

In a pre-industrial Malthusian world, the pace of technological advancement is so slow that you’re essentially working with a fixed number of resources, and every new mouth to feed means less to go around. But we don’t live in that world anymore. In my lifetime alone, we’ve gone from iPods to iPhones to ChatGPT. And to make those advances, firms needed investment in research and development.

Importantly, there is both a supply and demand side to investment. The supply side comes from savings: my paychecks are directly deposited in an account at Bank of America, and Bank of America uses those deposits to make loans.¹ Borrowers take those loans and use them to buy a house or build a new factory or start a business. But they only make these investments if there is demand for them: nobody is taking out loans to build a new iPod factory because people don’t use iPods anymore.

And slowing population growth is a double whammy to investment. On the demand side, smaller future generations mean lower expected demands for goods and services, which will make firms less willing to invest. And because people tend to save for retirement, an aging population means a larger supply of savings. What do you get when you combine diminished demand with surplus supply? Lower prices—in this case, a real neutral rate of interest, to use the technical term.

Alvin Hansen was really worried about this. 

He calculated that “the growth of population in the last half of the nineteenth century was responsible for about forty per cent of the total volume of capital formation in western Europe and about sixty per cent of the capital formation in the United States” and that if his math was even close to right, “it will be seen what an important outlet for investment is being closed by reason of the current rapid decline in population growth.”

This was Matt’s thesis in “Low interest rates are a curse — we need massive fiscal imprudence”—that Hansen was right, and slowing population growth leads to lower GDP growth and a lower neutral rate. Hansen’s prediction turned out wrong because he failed to predict the Baby Boom, but when you look at the data, it’s consistent with his theory. David Beckworth ran the numbers and found a positive correlation between real 10-year bond yields and population growth rates.

Matt wrote that “sensible people should think of the low interest rates as positively harmful” because they “simultaneously empower dangerous right populism and far-left utopianism, while eroding any incentive for people to think seriously about cost-benefit or spending prioritization” with deleterious effects for our political economy. He prescribed “massive fiscal imprudence” to get rates back up.

Well, Uncle Sam spent about $5 trillion on stimulus during the pandemic. The effective federal funds rate is now 5.33%. Per our recent polling over at Blueprint, voters are concerned about the deficit and government spending again. So, are we out of the woods? Given that Matt and I have precisely zero economics doctorates between the two of us, I decided to call up some experts. 

Experts say…

Astute readers will note that Hansen’s paper was published in 1939, which was a long time ago. Secular stagnation popped back into the macroeconomics discourse because of this 2013 IMF speech by Larry Summers. What is secular stagnation? Briefly, it describes a paradigm where the economy is stuck with low growth, low interest rates, high unemployment, and high debt, which makes it hard to get out of the hole using either monetary or fiscal policy.

Summers expanded on his argument in a 2016 Foreign Affairs essay, in which he contends that Robert Gordon and Ben Bernanke’s proposed explanations for the phenomenon—falling productivity growth and a “savings glut” from emerging markets, respectively—don’t match the data, and argues that expansionary fiscal policy is the optimal way to address it.

I spoke with Summers a few weeks ago to see how his views have changed since then. He told me that “the essence of secular stagnation is not some particular explanation; it’s the basic change in the fundamental factors driving savings relative to those driving investment,” pointing to a “combination of slower population growth and inequality.”

But he thinks that secular stagnation is “not the right paradigm now, post-pandemic and post-stimulus. “Economic theories are not like physics theories”, he said. “They fit for a period. So, I wasn’t contradicting myself when I went from worrying about secular stagnation to worrying about fiscal policy. I was responding to a radically changed world.”

Summers thinks that “while 2009 stimulus should’ve been larger, 2021 stimulus should’ve been smaller” and that a combination of overshooting on fiscal policy and “changing consumer technology and changing investment orientation” have altered the fundamentals. Specifically, he pointed towards a “substantial increase in green investment, substantial increase in resilience investment, prospectively a big increase in energy and compute investment related to AI” as possible channels for new investment.

For other views, I spoke to Matt Klein and Yale economists Giuseppe Moscarini, Michael Peters, and Marnix Amand about what they thought was the big driver of secular stagnation. 

Klein and Moscarini emphasize income distribution, while Peters and Amand emphasize aging — with Peters noting that aging means "falling business dynamism" and Amand citing lower rates of firm creation. Moscarini and Klein are more focused on the balance between saving and consumption, with Moscarini emphasizing that the rich have higher savings rates while Klein says "people spend their Social Security checks on stuff!" 

Are we out of the woods yet?

If new channels for investment have driven up demand, then we would expect the neutral interest rate to go up. Pre-pandemic, 2.5% was probably higher than the neutral rate; this can be inferred from the small slowdown in the labor market that happened when Janet Yellen inched rates up in 2015. 

Today, the neutral rate is probably higher than 2.5%, considering America’s roaring labor market and robust economic growth with much higher nominal and real rates than in 2015 and 2016. 

Yellen thinks the jury is still out on whether the neutral rate has been lifted, and Jay Powell told the Senate Banking Committee that “rates rights now are well into restrictive territory” and “well above neutral.” But Summers thinks it has gone up: he recently said on Bloomberg TV that the (nominal) neutral rate is much higher than 2.5% and likely over 4%. In a separate segment, he said that the most recent jobs report “fits the thesis that the neutral rate is much higher than people suppose.”

It’s worth stepping back at this point to consider why interest rates go up or down. Generally, it costs more to borrow for longer periods than shorter ones, because investors demand a term premium for locking their money up for longer. Similarly, illiquid investments must pay out a liquidity premium to lenders. (This is how David Swensen grew the Yale endowment from $1 billion to $30 billion.) And a risky loan carries a higher interest rate than a safe one. Research by Paul Schmelzing, Ken Rogoff, and Barba Rossi has found that real bond yields have been steadily falling since 1311. My guess is that a large part of this drop can be chalked up to falling risk premia, as society has become less violent and develops “institutions that let people do business with strangers because they can rely on rule of law,” as Summers put it.

But now, markets seem to expect that rates will remain elevated, at least for the next few years. That puts governments between a rock and a hard place. Higher rates mean the end of fairytale budgeting, where the government can increase social spending and defense spending and cut taxes and keep everyone happy. And in the long run, Summers believes that we have “a chronic debt problem.” 

“Government provides for national security, invests in science, supports the aged, and deals with inequality and its various consequences”, he told me. “And all of those things have become more necessary relative to other things than has been true historically. And the relative price of the kinds of things government buys, like education and healthcare, have gone up relative to the kind of things that people buy, like television sets and shirts. Therefore, the scale of government is likely to necessarily increase, and we will need more tax revenue.”

Even if the scale of government does not increase, he added, we will need more revenue, because Baumol’s cost disease will make it more expensive to maintain even the current level of services.

Summers said there is “a substantial chance we’ll have to go beyond the top 1%” to raise the necessary revenue. He identified as starting points “increased enforcement of tax laws we already have to close the tax gap, repealing the Trump tax cuts, raising the corporate tax to 26%, and increasing the capital gains tax.” In the long run, “the only way to produce durable improvements in the economy is on the supply side”, which is why he thinks that Matt’s argument that the focus should now shift towards things like NEPA reform and cutting zoning regulations is correct. “Our problem from 2013-2020 was chronic slack. Extra demand moved people from unemployment to employment; today we’d be moving people from one activity to another, and we have to think about the relative benefit of the two different activities.”

A new macroeconomic paradigm, if we can keep it

I found Summers’ argument that the combination of the green energy transition, great power competition with China, and a general backlash to free trade has opened new demand channels for investment persuasive. (I am a bit more skeptical of the argument around AI but will also admit that this is not a space I follow particularly closely.) If those channels boost investment demand, that will ultimately push the neutral rate up, giving central banks more firepower when the economy needs stimulus. 

But regarding the primary driver of secular stagnation, I found Peters, Amand, and Hansen more persuasive. Which worries me. To the extent that the problem is changes in the income distribution, well, politics ain’t beanbag, but on paper the solution is straightforward: redistribution. Population aging is trickier. Yes, you can buy yourself time by increasing immigration, but that’s politically dicey—more so in my professional opinion than redistribution. And ultimately it only delays the problem, because immigrants have to come from somewhere, and the global population is expected to peak around 2086, when I’d still be relatively young for a senator.

If the link between population aging and productivity growth exists, and I think it does, then some degree of secular stagnation is already baked in. As Hansen wrote in 1939, “We are thus rapidly entering a world in which we must fall back upon a more rapid advance of technology than in the past if we are to find private investment opportunities adequate to maintain full employment.”

Maybe green tech, reshoring, and remilitarization will provide those opportunities. Summers thinks they will. “If I had to guess, secular stagnation was a phase, and just like the people who thought it would return after World War II turned out to be wrong, those who think it will return after the pandemic turns out to be wrong. But this is quite uncertain.” Olivier Blanchard thinks otherwise, though, arguing that the decline in real rates on government bonds “has been a steady and global one since the 1980s” which “neither COVID nor inflation have done anything to reverse.” To the extent that higher incomes and older populations are driving rates down, he sees “higher rates as an interlude.”

Importantly, both Summers and Blanchard acknowledge there is substantial uncertainty around how much investment will go up. Ultimately this is more a question of politics than economic theory. On paper, it is possible to invest our way out of the low-rate trap; in practice, governments might simply not do that. As Amand argues, “if secular stagnation is still a problem five years from now, we haven’t found any worthwhile investment opportunities and then we haven’t solved far worse issues.”

The future is hard to predict, and I can tell you two plausible stories: one where new channels for investment break us out of the secular stagnation paradigm, and one where demographic aging pulls us back into it. If I had to guess, I’d say that the second story is more likely to be true. 

I realize that this is a very wishy-washy answer to my original question, but you’ll have to wait until I write my thesis for a more detailed one.

---

Milan Singh is an intern at the Niskanen Center; a partisan hack at Blueprint; and a staff columnist at the Yale Daily News. He is also a sophomore studying Economics and History at Yale and was previously Slow Boring’s researcher.

1. Yes, I’m aware that a commercial bank making a loan is money creation if you want to be precise about it.