Hacker News Reader

Comment by sgerenser | Hacker News Reader

"The Consumer Price Index for All Urban Consumers (CPI-U) rose 0.1 percent in November on a seasonally adjusted basis, after increasing 0.4 percent in October, the U.S. Bureau of Labor Statistics reported today. Over the last 12 months, the all items index increased 7.1 percent before seasonal adjustment."

I feel like the rise of only 0.1% October->November is more significant than the 7.1% YOY headline number, which is taking into account a lot of inflation that has already happened months ago. Of course, back in July we had a 0.0% MoM CPI increase that looked promising, but that didn't last long: https://www.bls.gov/news.release/cpi.nr0.htm

View on HN

The MoM numbers are important, but the breakdown is also important. The breakdown is worrisome - MoM declines were almost entirely driven by gas and other oil-related commodities, and we're still seeing very significant inflation in food (0.5% MoM) and shelter (0.6% MoM). Food and energy are transient (prices can go down as easily as they go up), but shelter and wages are sticky (they very rarely go down, and usually only in severe recessions.

The anatomy of the 1970s recession was similar - inflation in oil had stopped by 1975, but the worst of the inflation was carried through 1977-1981 by mortgage rates, rents, and wages.

Great detail, thanks. Any suggestions for further analysis/reading on the 70s recessionary environment?

Didn’t you have the black swan of the US going off the Gold Standard as a major inflationary driver at that time? Is there a similar looming inflationary pressure you see now?

My (limited) model is we’re seeing some supply-side disruptions and a some hot demand from Covid stimulus, but it’s not clear to me if these drivers are likely to continue through 2023 and beyond. (I assume mostly not, with Ukraine being the big problem, but mostly for EU rather than US.)

My favorite paper on the subject is:

https://www.nber.org/system/files/chapters/c11462/c11462.pdf

I'd also encourage you to look at actual numeric data from the time period - measured CPI [1] across different categories, fed funds rate [2], money supply, etc.

The data tells a very different story from the story - for one, it was neither caused by oil shocks, nor limited to the 1970s. It actually started in 1968, and the 1973 oil shock happened when the U.S. was already in recession from a Fed tightening that began in 1972 to deal with 1970's high inflation. Personally I'd attribute the cause as a series of poor decisions that were papered away by low interest rates, but which eventually compounded to devalue the currency. Vietnam took many young Americans out of the workforce and redirected production to war, Nixon pressured his Fed chief to lower interest rates, Nixon took us off the gold standard, Nixon introduced price controls (which further compounded supply issues), the oil shock hit, banks raised interest rates to compensate for inflation, which raised the cost of housing, which caused more inflation, until Volcker finally caused a massive recession and got it under control.

Note also that there were multiple waves of inflation (6.2% @ 1969, 12.3% @ 1974, 13.3% @ 1979) + Fed tightening (9% @ 1970, 11% @ 1972, 13% @ 1972, 18% @ 1980, 19% @ 1981). These were effective but not persistent - in between inflation fell to 3.3% @ 1971 and 4.9% @ 1976. Even in very high-inflation years you had some months with virtually no inflation - for example July 1973 (0.1% MoM), March 1974 (0.2% MoM), July 1980 (0%).

History doesn't repeat itself, but it rhymes. IMHO this was caused by having an economy that's very tightly optimized for ZIRP & globalization; introducing a pandemic that killed a million Americans, took another ~4.5M out of the workforce, and closed borders; and adding on some geopolitical black swans like the Ukraine war. Now workers need to reallocate from speculative high-margin activities like tech startups back to fundamentals like growing food and hauling trucks, and that is unlikely to happen unless the wages for truck drivers in the future exceed those of software developers now. We'll get cycles in between as the Fed tightens and loosens and causes recessions, but we don't fix the root problem until average income is ~$200K/year.

[1] https://www.usinflationcalculator.com/inflation/consumer-pri...

[2] https://fred.stlouisfed.org/series/FEDFUNDS

Shelter and wages are lagging indicators -- they go up months after the impetus for their increase.

Oil prices doubled between 1975 and 1980 so it was still contributing.

That's the problem - aside from only being sampled every 6 months, shelter reflects what people are actually paying, which includes a large number of people who haven't renewed their rents or who paid a lot less for their home and have no idea what it'd rent for now. As these people renew their leases or seek new housing, they're going to have to pay market rents, which in many cases have gone up a lot more than the measured CPI component. That bakes in a large future increase in the CPI, simply from price increases that have already happened.

The YoY that they report is definitely not useful during slight inflation like we're having. I wish the headlines were more about the month to month because I think people see the number and freak out that prices just went up 7.1 percent.

I think if MoM were widely reported, it would be stated as annualized to make sense for people in a headline. This would cause a lot of overreaction.

> I feel like the rise of only 0.1% October->November is more significant than the 7.1% YOY headline number

You are correct.

They always report the much more meaningless headline number. The current inflation rate is around 1.2%, well below Powell's target.

Of course we've already overshot, and are still at risk of a manufactured recession. Which was the goal, workers were getting uppity and We Can't Have That.

Reply on news.ycombinator.com