Good piece. One thing about the last chart, from the Fed note, which reinforces Jared's unwillingness to agree that it's a slam dunk that we are in a new high-inflation regime. Yes, the diffusion indices may be showing that more prices are rising by at least 3% annualised in the latest 6-month period than during 2014-19, not only in the US but in the euro area, the UK and Canada. But in ALL those economies inflation during 2014-19 was below the central bank's target: in the 6 years to Q4 2019, headline inflation averaged 0.9% in the euro area, 1.5% in the UK, 1.8% in Canada and 1.3% in the US. So SOME shift in the diffusion of price increases towards the higher end is consistent with inflation being on target, instead of below. And indeed the latest pre-Iran-war inflation data for the euro area and Canada were consistent with that benign interpretation: 1.9% for the euro area in Feb and 1.8% for Canada. The UK was a somewhat different story, with Feb inflation at 3%.
The UK and US have both damaged their economies by buying anti-trade populist propaganda. The UK stupidly chose Brexit which is helping keep their inflation higher and Trump imposed tariffs and now is driving up oil prices with his insane war and blockade of the Strait of Hormuz.
That's been happening. A standard econ analysis might suggest this puts downward pressure on inflation because it leads to slower wage growth (and wage growth feeds into price growth). I wouldn't go there and think about this development more as an inequality phenomenon than an inflationary one.
But has corporate concentration been increasing in Canada, the UD, and the Euro Area as well? (Seem unlikely to me but I don’t have any data) This looks to be a phenomenon not uniquely American.
Inflation has been inexorably linked to the supply of oil . The current spate of inflation is no exception - although Trump's bone-headed tariffs aided and abetted his (and Netanyahu's) even crazier attack on Iran. And, even though solar and wind promise to reduce the dependence of inflation on the supply of oil, Trump predictably opposes wind energy and wants to "drill, baby, drill!" But Trump too shall pass. Given the rapid adoption of renewables in the world's energy mix, and the dizzying pace of technological innovation, only maladminstration comparable to Trump's could fail to put a dent in the inflation rates of the future.
Inflation is a complex issue and while in limiting cases, eg. true supply shocks, direct attribution is somewhat reasonable, but beyond that the true answer is inflation is a complex issue with multiple interacting components. Simple explanation will always fall short because inflation is essentially determined by human psychology and resulting behavior. In any even partially developed society a large fraction of production and consumption is driven by wants and not needs. But a more important factor is money. Even a smart economist like Krugman in a recent Substack piece wrote something a good high school level science student knows is wrong: that money is both a medium of exchange and a store of value. This is impossible, both by dimensional analysis and basic thermodynamics. Money has zero real value so it can’t be a store of value. In other language you can’t eat money to survive (eg. take the limit of the price of food as the food supply goes to zero). The store of value component is a combination of an implicit contract that others in the future will accept your coupons and the supply of coupons at any point in time relative to underlying real economic parameters. The pandemic inflation demonstrated the factors clearly. People had more coupons given out by the government but there was less real output, but still far above the needs level. People had coupons (dollars) and wants and so prices increased. The coupons are created by banks when they create a new loan. Monetary policy is a poor tool to manage the supply of coupons for multiple reasons: it is very indirect, it slightly changes the math for banks who create money, the effects are indirect, the effects have a long time constant. A scientist would never design a system like this to control something. Control of inflation is, scientifically, simple, using the classic PID algorithm. The government can, continuously, slightly vary tax and spending levels to keep inflation at any set point. It is the way the thermostat in your house works or how the most complex systems in the world, like the tools that make semiconductor chips, or fighter aircraft. It works everywhere. We choose not to control something we could easily control.
A huge lesson was learned from the pandemic era supply chain snarls: the just-in-time logistics model was a high wire act without a safety net. Switching to a just-in-case model doesn’t come without a price; redundancies and inventory costs have to show up somewhere. Granted, this change ought to have shown up as a one-time inflationary hit — and maybe it did, but spread out over a number of years as different firms in different sectors adopted just-in-case strategies at different rates.
Then there’s the service sector. My gut tells me that any decrease in service volumes means fixed costs have to be recovered from a smaller customer base.
"..it’s not so much inflation—the rate of price changes—that’s gotten deeply under their skin. It’s the elevated price levels.."
While I agree with this statement, I can't help but focus on the period 1983 to 2000 when (with a couple of exceptions) we experienced one of the longest periods of falling inflation and interest rates in recent history. While costs cumulatively increased during this time period, the cost of any debt used to purchase goods and services steadily fell, which would have tended to mask the effect of these cost increases. Similarly, the increase in wealth for Baby Boomers, and generally for the top 50% of wealth holders, during this same time period would have tended to offset the cumulative effects of rising costs. As you note, however, the Pandemic brought about both a sharp increase in inflation, as well as an end to the 25+ year run of falling interest rates. With no where to hide the effects of rising costs, many consumers have good reason to be grumpy.
Demographic contexts? US population grew over the X axis of these charts ( coincides w/ baby boom) but these days the make up of X is changing. Less & less births, the huge slug of us boomers departing. Sentiment is both an individual and cohort-reflective value, no? Maybe this is yet another and final expression of the boom. Miss us yet?
1) Quibbling slightly on what graph #2 seems to suggest if inflation manages to hold at 3%, once the covid spike passes, in 2 or 3 years, the cumulative 5-year inflations would drop a bit, roughly to (1.03)^5-1 = .16.
(If I understand what is plotted;
inflation holding at 3% of course seems highly unlikely based on oil, gas, and fertilizer from the Hormuz area, according to all analysts I have followed.)
2) Yes, the lecture later in the week with the graphs I know about. At least if it is the one I've seen advertised in my local newspaper, which is the main local paper in that area. (The Gamble Memorial Lecture at UMASS Amherst)
There were 2 big half-page adds promoting it last week, and a big full-page one yesterday. (It's a normal-sized newspaper--not a tabloid. The ad was really big! I don't know if I've seen a full-page ad in it before!)
I can't find the ad outside of a paywall, but it was something like this UMASS anouncement:
(Everyone: Note the picture of J.B., which was also in the giant ad. The same picture, with the "wait a minute, buddy" gesture, apparently directed at some reporter tossing out an inaccurate idea falsely besmirching Biden econ policy, used to indentify this substack.)
Good luck, J.B., with the talk. (I have not quite enough energy in my old age these days to trek the 11 miles to see it in person, but I will certainly catch the video of it, which will undoubtedly be on youtube.)
Nice piece. One nit-pick: why on earth is the vertical axis on your second chart, the one with the foothill/half mountain, denominated in monthly changes? The language accompanying it suggests cumulative 5-year changes. It would help readers who are less geeky than I am when it comes to extracting messages from charts to “clear the graph” (explain how to read the chart).
It's five year changes ending the month of the x-axis. Will think about how to make that clearer. I need to take more time with such things but tend to bang these posts out in a frenzy!
Agreeing with Geoff. I don't see a problem with the graph.
(Also, if my hunch is correct that the graph will be used in a lecture at UMASS Amherst--the big UMASS flagship--in a few days, then the vertical scale being in proportion rise, so that .2 means prices went up 20% over the prior 5 years, would be a nice little quantitative critical thinking exercise for the young students at UMASS.
Hopefully, in fact, after the lecture, during the Q & A, some student will make a comment needing the explanation of "cumulative proportion rise", perhaps as well "monthly evaluation" of same rise, and lots of kids will learn from it! (Possibly some professors, as well, in non-economics disciplines!)) (😊)
Not really sure (re your AI question). A lot of modeling is pretty theoretic and, while I'm sure AI applies, it's probably ancillary. But the really interesting development I'm seeing is the use of AI for econ forecasting. That could turn out to be useful/important.
Good piece. One thing about the last chart, from the Fed note, which reinforces Jared's unwillingness to agree that it's a slam dunk that we are in a new high-inflation regime. Yes, the diffusion indices may be showing that more prices are rising by at least 3% annualised in the latest 6-month period than during 2014-19, not only in the US but in the euro area, the UK and Canada. But in ALL those economies inflation during 2014-19 was below the central bank's target: in the 6 years to Q4 2019, headline inflation averaged 0.9% in the euro area, 1.5% in the UK, 1.8% in Canada and 1.3% in the US. So SOME shift in the diffusion of price increases towards the higher end is consistent with inflation being on target, instead of below. And indeed the latest pre-Iran-war inflation data for the euro area and Canada were consistent with that benign interpretation: 1.9% for the euro area in Feb and 1.8% for Canada. The UK was a somewhat different story, with Feb inflation at 3%.
The UK and US have both damaged their economies by buying anti-trade populist propaganda. The UK stupidly chose Brexit which is helping keep their inflation higher and Trump imposed tariffs and now is driving up oil prices with his insane war and blockade of the Strait of Hormuz.
What about increased corporate concentration and a rising profit share?
That's been happening. A standard econ analysis might suggest this puts downward pressure on inflation because it leads to slower wage growth (and wage growth feeds into price growth). I wouldn't go there and think about this development more as an inequality phenomenon than an inflationary one.
But has corporate concentration been increasing in Canada, the UD, and the Euro Area as well? (Seem unlikely to me but I don’t have any data) This looks to be a phenomenon not uniquely American.
Please note it is not the Ukrane war, it is Russias invasion war on Ukraine. Please change the text to reflect the reality.
Inflation has been inexorably linked to the supply of oil . The current spate of inflation is no exception - although Trump's bone-headed tariffs aided and abetted his (and Netanyahu's) even crazier attack on Iran. And, even though solar and wind promise to reduce the dependence of inflation on the supply of oil, Trump predictably opposes wind energy and wants to "drill, baby, drill!" But Trump too shall pass. Given the rapid adoption of renewables in the world's energy mix, and the dizzying pace of technological innovation, only maladminstration comparable to Trump's could fail to put a dent in the inflation rates of the future.
Inflation is a complex issue and while in limiting cases, eg. true supply shocks, direct attribution is somewhat reasonable, but beyond that the true answer is inflation is a complex issue with multiple interacting components. Simple explanation will always fall short because inflation is essentially determined by human psychology and resulting behavior. In any even partially developed society a large fraction of production and consumption is driven by wants and not needs. But a more important factor is money. Even a smart economist like Krugman in a recent Substack piece wrote something a good high school level science student knows is wrong: that money is both a medium of exchange and a store of value. This is impossible, both by dimensional analysis and basic thermodynamics. Money has zero real value so it can’t be a store of value. In other language you can’t eat money to survive (eg. take the limit of the price of food as the food supply goes to zero). The store of value component is a combination of an implicit contract that others in the future will accept your coupons and the supply of coupons at any point in time relative to underlying real economic parameters. The pandemic inflation demonstrated the factors clearly. People had more coupons given out by the government but there was less real output, but still far above the needs level. People had coupons (dollars) and wants and so prices increased. The coupons are created by banks when they create a new loan. Monetary policy is a poor tool to manage the supply of coupons for multiple reasons: it is very indirect, it slightly changes the math for banks who create money, the effects are indirect, the effects have a long time constant. A scientist would never design a system like this to control something. Control of inflation is, scientifically, simple, using the classic PID algorithm. The government can, continuously, slightly vary tax and spending levels to keep inflation at any set point. It is the way the thermostat in your house works or how the most complex systems in the world, like the tools that make semiconductor chips, or fighter aircraft. It works everywhere. We choose not to control something we could easily control.
A huge lesson was learned from the pandemic era supply chain snarls: the just-in-time logistics model was a high wire act without a safety net. Switching to a just-in-case model doesn’t come without a price; redundancies and inventory costs have to show up somewhere. Granted, this change ought to have shown up as a one-time inflationary hit — and maybe it did, but spread out over a number of years as different firms in different sectors adopted just-in-case strategies at different rates.
Then there’s the service sector. My gut tells me that any decrease in service volumes means fixed costs have to be recovered from a smaller customer base.
Paul Volker may have had to lower the boom in the late 70’s and early 80’s
by increasing interest rates to finally end the run away inflation .
But if Arthur Burns had not accommodated Nixon’s desire for
a booming economy by keeping monetary policy - easy -
And - if we hadn’t abandoned the international gold standard in August
of 71’ . The dollar would not have dropped as precipitously as it did
and the Volker cure might not have
had to be as extreme as it was .
But it was - and we suffered .
I’m not longing for the possible deflationary experiences of the
2008 - 10 Great Recession either .
It just seems like however long it takes to sober up - we forget about the hang over - and go on another binge
again .
"..it’s not so much inflation—the rate of price changes—that’s gotten deeply under their skin. It’s the elevated price levels.."
While I agree with this statement, I can't help but focus on the period 1983 to 2000 when (with a couple of exceptions) we experienced one of the longest periods of falling inflation and interest rates in recent history. While costs cumulatively increased during this time period, the cost of any debt used to purchase goods and services steadily fell, which would have tended to mask the effect of these cost increases. Similarly, the increase in wealth for Baby Boomers, and generally for the top 50% of wealth holders, during this same time period would have tended to offset the cumulative effects of rising costs. As you note, however, the Pandemic brought about both a sharp increase in inflation, as well as an end to the 25+ year run of falling interest rates. With no where to hide the effects of rising costs, many consumers have good reason to be grumpy.
Demographic contexts? US population grew over the X axis of these charts ( coincides w/ baby boom) but these days the make up of X is changing. Less & less births, the huge slug of us boomers departing. Sentiment is both an individual and cohort-reflective value, no? Maybe this is yet another and final expression of the boom. Miss us yet?
1) Quibbling slightly on what graph #2 seems to suggest if inflation manages to hold at 3%, once the covid spike passes, in 2 or 3 years, the cumulative 5-year inflations would drop a bit, roughly to (1.03)^5-1 = .16.
(If I understand what is plotted;
inflation holding at 3% of course seems highly unlikely based on oil, gas, and fertilizer from the Hormuz area, according to all analysts I have followed.)
2) Yes, the lecture later in the week with the graphs I know about. At least if it is the one I've seen advertised in my local newspaper, which is the main local paper in that area. (The Gamble Memorial Lecture at UMASS Amherst)
There were 2 big half-page adds promoting it last week, and a big full-page one yesterday. (It's a normal-sized newspaper--not a tabloid. The ad was really big! I don't know if I've seen a full-page ad in it before!)
I can't find the ad outside of a paywall, but it was something like this UMASS anouncement:
https://www.umass.edu/news/article/former-white-house-council-economic-advisers-chair-jared-bernstein-present-2026-umass
(Everyone: Note the picture of J.B., which was also in the giant ad. The same picture, with the "wait a minute, buddy" gesture, apparently directed at some reporter tossing out an inaccurate idea falsely besmirching Biden econ policy, used to indentify this substack.)
Good luck, J.B., with the talk. (I have not quite enough energy in my old age these days to trek the 11 miles to see it in person, but I will certainly catch the video of it, which will undoubtedly be on youtube.)
Nice piece. One nit-pick: why on earth is the vertical axis on your second chart, the one with the foothill/half mountain, denominated in monthly changes? The language accompanying it suggests cumulative 5-year changes. It would help readers who are less geeky than I am when it comes to extracting messages from charts to “clear the graph” (explain how to read the chart).
It's five year changes ending the month of the x-axis. Will think about how to make that clearer. I need to take more time with such things but tend to bang these posts out in a frenzy!
Possibly Ed was reading the vertical scale as 0.2 means 0.2% (looking like a monthly inflation rate)? Whereas it actually means a 20% 5-year change.
Agreeing with Geoff. I don't see a problem with the graph.
(Also, if my hunch is correct that the graph will be used in a lecture at UMASS Amherst--the big UMASS flagship--in a few days, then the vertical scale being in proportion rise, so that .2 means prices went up 20% over the prior 5 years, would be a nice little quantitative critical thinking exercise for the young students at UMASS.
Hopefully, in fact, after the lecture, during the Q & A, some student will make a comment needing the explanation of "cumulative proportion rise", perhaps as well "monthly evaluation" of same rise, and lots of kids will learn from it! (Possibly some professors, as well, in non-economics disciplines!)) (😊)
If we had a sane prez, except for you datawonkers, care?
We’re all waiting for a catastrophic economic event, which thankfully may never happen.
If the Ds win in November, especially a “double play,” will anyone actually care if inflation is 3 instead of 2%?
BTW, how is AI impacting economic modeling?
Not really sure (re your AI question). A lot of modeling is pretty theoretic and, while I'm sure AI applies, it's probably ancillary. But the really interesting development I'm seeing is the use of AI for econ forecasting. That could turn out to be useful/important.