The Great Recession and its aftermath sparked a lot of rethinking about economics and economic policy. One recurring theme in this rethinking is the danger of debt.

In “This Time Is Different,” a recent book that explores financial crises throughout history, Harvard University economists Carmen Reinhart and Ken Rogoff show how run-ups in debt are key to setting up crises. But what kind of debt should set off alarms for economists and policymakers? A new working paper by economists Atif Mian and Emil Verner of Princeton University and Amir Sufi of the University of Chicago points to household debt as a source of instability and lower economic growth.

The working paper looks at the relationship between household debt and economic growth in 30 countries from 1960 to 2012. Earlier research found that recessions driven by household debt—particularly mortgage debt—are particularly nasty. But Mian, Sufi, and Verner push this research a bit further to find that high levels of household debt, measured as the ratio of household debt to gross domestic product, predict lower economic growth.

The three economists can’t say exactly why household debt has a negative effect on growth and stability. Instead they point out that their results are consistent with models of the economy where the supply of credit expands, lenders are more likely to lend, or a decline in risk allows households to take out more loans to finance consumption.

The dynamics of how household debt affects economic growth are quite interesting. Higher debt results in more consumption by households and a larger share of economic output coming from consumption. At the same time, this results in the country running a larger current account deficit with the rest of the world as imports increase, with consumption goods making up a larger share of those imports. Once the economy enters a recession, however, imports fall dramatically and the country starts to export more than it imports. This mechanism allows the country that ran up debt to bounce back by selling goods to the global economy.

But that mechanism only works if the rest of the global economy is doing well and other countries didn’t also see an increase in the household debt. If the entire global economy has more household debt, then individual countries will have nowhere to export their goods when growth stalls. Mian, Sufi, and Verner find that an increase in global household debt is also predictive of lower economic growth for individual countries, and that countries particularly dependent on trade are affected even more by this global buildup.

A lot of these dynamics sound a lot like what happened during the 2000s in the run up to the Great Recession of 2007-2009, so it’s plausible that the general results are artifacts of what happened over the past 15 years and are not symptomatic of a more systemic issue. But when Mian, Sufi, and Verner look at the data before 2000, they find very similar results. And using those results, they can show that the run-up in household debt during the 2000s indicated a recession of the size and severity of the Great Recession.

These results seem almost obvious given what’s happened over the past decade, but the idea that household debt is bad for economic growth runs counter to quite a few economic models. Mian, Sufi, and Verner show that both the International Monetary Fund and the Organisation for Economic Co-Operation and Development have consistently been overly optimistic about economic growth because they have seen increasing household debt as a positive for economic growth. The evidence now points to just the opposite.

These results are important not for shaming certain institutions, but rather to show how important it is to consider the role of household debt in economic stability and growth.