Larry Summers claims he can prove inflation is way closer to the 1970s than people think—and that a deep recession may be the only way to end it

Last year, months before the Federal Reserve began raising interest rates to combat inflation, former Treasury Secretary Larry Summers warned that inflation was already a problem, and would get a lot worse in 2022.

His predictions were accurate, with inflation now sitting at a nearly 40-year high of 8.3%. But now Summers is doubling down, cautioning that inflation may be worse than official numbers show, and that a massive economic contraction may be needed to rein in prices.

In a new paper published by the National Bureau of Economic Research, a nonprofit and nonpartisan economic research organization, Summers and his coauthors claim to have devised a new methodology to analyze inflation that they say is more accurate than the consumer price index (CPI), the most widely used yardstick. And what it finds is that taming today’s inflation might require a much more aggressive Federal Reserve.

“Using these series, we find that current inflation levels are much closer to past inflation peaks than the official series would suggest,” the authors wrote.

Summers and the paper’s coauthors, economists Judd Cramer and Marijn Bolhuis, write that the new metric better analyzes modern spending patterns than the CPI.

It isn’t the first time the CPI has come under fire from critics, who have in the past argued that the index tends to underestimate real inflation rates.

The Bureau of Labor Statistics, which calculates the monthly CPI, measures the annual change in prices among over 200 categories of commonly purchased goods and services. But the authors argue that this approach can lead to a misrepresentation of the data, as flaws in past inflation measurement methodologies led to inaccurate readings.

The paper argues that prior to 1983, the CPI did not correctly account for consumer spending on housing, including both the final home purchase value and the value of mortgage rates spread out over 30 years. The pre-1983 index included both home purchase prices and the total outlay of mortgage payments, despite mortgages being paid out gradually over several years.

“Including both led to a larger share of housing in the consumption basket,” the authors wrote. This metric was in use between 1953 and 1983, when the Bureau of Labor Statistics “stripped away the investment aspect of housing” and changed the CPI to “isolate owner-occupiers’ consumption of residential services.”

The authors wrote that the way housing spending was measured pre-1983 was “without conceptual foundation,” and that the methodology “resulted in a substantial upward bias in the CPI.”

Summers and colleagues argue that this caused inflation measures before 1983 to look “artificially high at the beginning of the tightening cycle,” and to recede “artificially fast.”

According to Summers’ methodology, today’s inflation rate is dangerously close to that of another period of economic distress in U.S. history: the early 1980s. In 1980, the official inflation rate was 13.6%, although Summers’ index puts it at 9.1% when the alleged problem with housing spending is removed.

If Summers is right, his index would have huge implications for how far the Federal Reserve may have to go to fix inflation. It could mean that the Fed would have to resort to the same extreme means employed by the central bank to tame inflation in the 1980s under former Chairman Paul Volcker.

In 1980, Volcker authorized an increase in interest rates to the tune of 20% to combat inflation, which plunged the economy into a severe recession that lasted until 1982 in the U.S. At the time, this was the worst economic downturn since the Great Depression. Indeed, the unemployment rate reached 10.8% in 1982, a high that wouldn’t be surpassed until the pandemic-fueled recession hit businesses in 2020.

Despite the recession that ensued, Volcker’s policy was seen as a win for advocates of an aggressive Federal Reserve and hawkish monetary policy in times of high inflation. But if Summers is right, it would mean that inflation may be significantly less responsive to monetary tightening than first believed, and that the current Fed administration under Chairman Jerome Powell may have to be even more aggressive to get results, despite the recession risks.

The Fed has been progressively raising interest rates since March in a bid to slow down the economy. And while last month’s inflation rate was slightly below its 8.5% peak in April, some economists, including Summers, believe prices will stay high for a while.

This story was originally featured on Fortune.com