Last Friday the Supreme Court overturned the doctrine of Chevron deference as part of its ruling in Loper Bright Enterprises v Raimondo. This might not have even been their most discussed ruling of the past week, but in my (non-lawyerly) opinion, there is a good chance it will be their most economically impactful ruling of the past decade. SCOTUSblog explains the basics:
the Supreme Court on Friday cut back sharply on the power of federal agencies to interpret the laws they administer and ruled that courts should rely on their own interpretation of ambiguous laws. The decision will likely have far-reaching effects across the country, from environmental regulation to healthcare costs.
By a vote of 6-3, the justices overruled their landmark 1984 decision in Chevron v. Natural Resources Defense Council, which gave rise to the doctrine known as the Chevron doctrine. Under that doctrine, if Congress has not directly addressed the question at the center of a dispute, a court was required to uphold the agency’s interpretation of the statute as long as it was reasonable. But in a 35-page ruling by Chief Justice John Roberts, the justices rejected that doctrine, calling it “fundamentally misguided.”
Justice Elena Kagan dissented, in an opinion joined by Justices Sonia Sotomayor and Ketanji Brown Jackson. Kagan predicted that Friday’s ruling “will cause a massive shock to the legal system.”
When the Supreme Court first issued its decision in the Chevron case more than 40 years ago, the decision was not necessarily regarded as a particularly consequential one. But in the years since then, it became one of the most important rulings on federal administrative law, cited by federal courts more than 18,000 times.
The most common reaction I’ve seen is that people expect this to reduce the power of executive branch agencies, both in general and relative to courts and businesses, likely resulting in deregulation. Thus those on the economic left have been mostly decrying the decisions, while free–marketers and businesspeoplehave mostly beencelebrating:
Back in January I encouraged you to follow the money in the Presidential race, by which I meant follow the betting markets. I suggested this was a good way to cut through the sometimes inaccuracy of polls, and the uncertainty of listening to any one expert or group of experts. Bettors in prediction markets can take all of these into account.
Lately of course the big question in the Presidential race is whether Biden will actually be the Democratic nominee. There is much uncertainty right now, and you will all kinds of predictions from experts, media quoting “inside sources,” and other such rumors. How are you, as a relatively uninformed outsider, supposed to know who to trust?
The answer again I will suggest is: watch the betting markets. And if you check the betting markets today (aggregated across multiple markets by EletionBettingOdds.com), you will see that Biden and Kamala Harris have roughly equal chances of becoming the next President (and Trump is about a 60% favorite):
We all know about inflation. One popular measure is the Consumer Price Index (CPI), which measures the change in price of a fixed basket of goods. The other popular measure used for inflation is the Personal Consumption Expenditures (PCE) price index. This index measures the price of what consumers actually purchase and captures the effects of consumers changing their consumption bundles over time. While the latter is a better measure for the prices at which consumers make purchases, it takes longer to calculate. In practice, the earlier CPI release gives a pretty accurate preview to the PCE price index.
While consumption is a substantial two-thirds of total expenditures in the US economy, other prices definitely matter. On average, a third of our income is spent on other things. Below is a stacked bar chart of quarterly GDP components – the classic Y=C+I+G+NX.* Investment spending composes a relatively stable 16.7% and Government spending composes about 16.5% of GDP. We almost never hear much about the price of these other things.
The Fed has now almost landed the plane, bringing us down from 9% inflation during the Covid era to something approaching their 2% target today. But it is not yet clear how hard the landing will be. Back in March I thought recurrent inflation was still the big risk; now I see the risk of inflation and recession as balanced. This is because inflation risks are slightly down, while recession risk is up.
Inflation remains somewhat above target: over the last year it was 3.3% using CPI, 2.7% by PCE, and 2.8% by core PCE. It is predicted to stay slightly above target: Kalshi estimates CPI will finish the year up 2.9%; the TIPS spread implies 2.2% average inflation over the next 5 years; the Fed’s own projections say that PCE will finish the year up 2.6%, not falling to 2.0% until 2026. The labels on Kalshi imply that markets are starting to think the Fed’s real target isn’t 2.0%, but instead 2.0-2.9%:
The Fed’s own projections suggest this to be the somewhat the case- they plan to start cutting over a year before they expect inflation to hit 2.0%, though they still expect a long run rate of 2.0%. In short, I think there is a strong “risk” that inflation stays a bit elevated the next year or two, but the risk that it goes back over 4% is low and falling. M2 is basically flat over the last year, though still above the pre-Covid trend. PPI is also flat. The further we get from the big price hikes of ’21-’22 with no more signs of acceleration, the better.
But I would no longer say the labor market is “quite tight”. Payrolls remain strong but unemployment is up to 4.0%. This is still low in absolute terms, but it’s the highest since January 2022, and the increase is close to triggering the Sahm rule (which would predict a recession). Prime-age EPOP remains strong though. The yield curve remains inverted, which is supposed to predict recessions, but it has been inverted for so long now without one that the rule may no longer hold.
Looking through this data I think the Fed is close to on target, though if I had to pick I’d say the bigger risk is still that things are too hot/inflationary given the state of fiscal policy. But things are getting close enough to balanced that it will be easy for anyone to find data to argue for the side that they prefer based on their temperament or politics.
To me the big wild card is the stock market. The S&P500 is up 25% over the past year, driven by the AI boom, and to some extent it pulls the economy along with it. The Conference Board’s leading economic indicators are negative but improving overall this year; recently their financial indicators are flat while non-financial indicators are worsening.
Overall things remind me a lot of the late ’90s: the real economy running a bit hot with inflation around 3% and unemployment around 4%; the Fed Funds rate around 5%; and a booming stock market driven by new computing technologies. Naturally I wonder if things will end the same way: irrational exuberance in the stock market giving way to a tech-driven stock market crash, which in turn pushes the real economy into a mild recession.
Of course there is no reason this AI boom has to end the same way as the late-90’s internet boom/bubble. There are certainly differences: the Federal government is running a big deficit instead of a surplus; there are barely a tenth as many companies doing IPOs; many unprofitable tech stocks already got shaken out in 2022, while the big AI stocks are soaring on real profits today, not just expectations. Still, to the extent that there are any rules in predicting stock crashes, the signs are worrying. Today’s Shiller CAPE is below only the internet and Covid meme-stock bubble peaks:
Again, this doesn’t mean that stocks have to crash, or especially that they have to do it soon; the CAPE reached current levels in early 1998, but then stocks kept booming for almost two years. I’m not short the market. But the macro risk it poses is real.
First, here is an updated chart on average wealth by generation, which gives us the first glimpse at 2024 data:
I won’t go into too much detail explaining the chart here, as I have done that in more detail in pastposts. But one brief explanatory note: I’m now labeling the most recent generation “Millennials & Gen Z (18+).” Because of the nature of the data from the Fed’s DFA, I can’t separate these two generations (it can be done with the Fed SCF data, but that is now 2 years old). This combined generation now includes everyone from ages 18 to 43 (which means that technically the median age is 30.5, not quite 31 yet), somewhere around 116 million people, which makes it a bit of a weird “generation,” but you work with the data you have. Note though that this makes the case even harder for young Americans to be doing well, as every year I am adding about 400,000 people to the denominator of the calculation, even though 18-year-olds don’t have much wealth.
What’s notable about the data is just how much the youngest “generation” in the chart has jumped up in recent years. They have now have about double the wealth that Gen X had at roughly the same age. Average wealth is about as much as Gen X and Boomers had 5-6 years later in life — and while there are no guarantees, odds are Millennial/Gen Z wealth will be much, much higher in another 5-6 years. You may notice at the tail end of the chart that Gen X and Boomers now have roughly equal amounts of average wealth at the same age (Gen X’s current age), while 2 years ago they were $100,000 ahead. I suspect this is just temporary, and Gen X will soon be ahead again, but we shall see.
Of course, the most common complaint about my data is that these are just averages, so they don’t tell us a lot about the distribution of wealth and could be impacted by outliers. That’s why I’m really excited to share this new data on wealth by decile from the 2022 Fed SCF survey. This data was put together by Rob J. Gruijters and co-authors, and it allows us to compare the wealth of Boomers, Gen X, and Millennials across the wealth distribution. You should read their analysis of the data, but in this post I’ll give my slightly different (and optimistic) interpretation of it.
For all three generations, wealth in the bottom 10% is negative when that generation is in their 30s. And for Millennials, it is the most negative: -$65,000 compared to -$30,000 for Gen X and -$17,000 for Boomers in the bottom decile (as always, the figures are adjusted for inflation). While I haven’t dug into the data, my suspicion is that student debt is driving a lot of the increase. Since this is households in their 30s, I suspect a lot of the bottom decile is composed of folks that just finished graduate and professional school, and are only now starting to acquire assets and pay down debt — they have very high earning potential, which means over their lifetime they will do great, but they are starting from behind. Again, we’ll have to wait and see, but I suspect many in the bottom will quickly climb up the wealth distribution over their working years.
That being said, in the following chart I have left off the bottom 10% for each generation, since displaying negative wealth would make the chart look a little weird. But this chart shows a very optimistic result: Millennials are doing better than Boomers across the distribution, and Millennials are ahead of almost all deciles for Gen X except a few, where they are essentially equal to Gen X (2nd, 7th, and 8th deciles).
The chart may be a little confusing (give me your suggestions to improve it!), but here’s how to read it. The blue bars show Millennial wealth relative to Gen X, at the same age, for each decile (excluding the bottom 10%). For example, the first bar shows that Millennials in the 2nd wealth decile had 100% of the wealth that Gen Xers in the 2nd wealth decile had at the same age — in other words, they were equal. The orange bars show Millennial wealth relative to Baby Boomer wealth at the same age, in the same decile (to repeat, it’s all adjusted for inflation).
Notice that other than the very first bar (Millennials vs. Gen X in the 2nd wealth decile), all of the other bars are over 100%, indicating that Millennials have more wealth than the two prior generations for almost every decile. For some of these, they are much, much greater than 100%. In the 5th decile (close to the median), Millennials have over 50% more wealth than Gen X and almost 200% (double the wealth) of the wealth of Boomers. That’s a massive increase!
A pessimistic read of the chart is that the biggest gains went to the top 10%. Though notice that’s only true relative to Baby Boomers. When compared with Gen X, the 4th and 5th deciles did better than the top 10% in terms of relative improvement. To relate this to the earlier chart in this post, it suggests that relative to Boomers, outliers at the top end might be skewing the average a bit, but that’s probably not the case relative to Gen X. And again, the broad-based gains are visible throughout the distribution from the 2nd decile on up.
Finally, on social media I’ve got several objections about the chart, such as folks not liking the log scale y-axis, and preferring the CPI-U for inflation adjustments instead of the PCEPI that I use. For those objectors, here is a different version of the chart:
Economics as a discipline really likes to boil things down to their essentials. There are plenty of examples. How many goods can one consume? Just two, bread and not bread. How can you spend your time? You can labor or leisure. How do you spend your money? Consume or save. It’s this last one that I want to emphasize here.
First, all income ultimately ends up being spent on consumption. Saving today is just the decision to consume in the future. And if not by you, then by your heirs. One determinant of inter-temporal consumption decisions is the real rate of return. That is, how many apples can you eat in the future by forgoing an apple eaten today? The bigger that number is, the more attractive the decision to save.
Further, since most saving is not in the form of cash and is instead invested in productive assets, we can also characterize the intertemporal consumption problem as the current budget allocation decision to consume or invest. The more attractive capital becomes, the more one is willing to invest rather than consume. The relative attractiveness between consumption and investment informs the consumption decision.
How attractive is investment? I’ll illustrate in two graphs. First, if the price of investment goods falls relative to consumption goods, then individuals will invest more. The graph below charts the price ratio of investment goods to consumption goods. Relative to consumption, the price of investment has fallen since 1980. Saving for the future has never been cheaper!
Of course, as in a price taker story, I am assuming that individuals don’t affect this price ratio. Truly, prices are endogenous to consumption/investment decisions. For all we know, it may be that the prices of investment goods are falling because demand for investment goods has fallen. But that doesn’t appear to be the case.
This week the University of the Arts in Philadelphia announced they were closing effective immediately, leaving students scrambling to transfer and faculty desperate for jobs. U Arts now joins Cabrini University and Birmingham-Southern as some the 20 US colleges closing or being forced to merge so far this year. This trend of closures is likely to accelerate given falling birth rates that mean the number of college-age Americans is set to decline for decades; short-term issues like the FAFSA snafu and rising interest rates aren’t helping either.
All this makes it more important for potential students and employees to consider the financial health of colleges they might join, lest they find themselves in a UArts type situation. But how do you predict which colleges are at significant risk of closing? One thing that jumps out from this year’s list of closures is that essentially every one is a very small (fewer than 2000 undergrad) private school. Rural schools seem especially vulnerable, though this year has also seen plenty of closures in major cities.
There appear to be a number of sourcestracking the financial health of colleges, though most are not kept up to date well. Forbes seems to be the best, with 2023 ratings here; UArts, Cabrini, and Birmingham-Southern all had “C” grades. If you have access to them, credit ratings would also be good to check out; Fitch offers a generally negative take on higher ed here.
In a 2020 Brookings paper, Robert Kelchen identified several statistically significant predictors of college closures:
I used publicly available data compiled by the federal government to examine factors associated with college closures within the following two to four years. I found several factors, such as sharp declines in enrollment and total revenue, that were reasonably strong predictors of closure. Poor performances on federal accountability measures, such as the cohort default rate, financial responsibility metric, and being placed on the most stringent level of Heightened Cash Monitoring, were frequently associated with a higher likelihood of closure. My resulting models were generally able to place a majority of colleges that closed into a high-risk category
The Higher Learning Commission reached similar conclusions. Of course, there is a danger in identifying at-risk colleges too publicly:
Since a majority of colleges identified of being at the highest risk of closure remained open even four years later, there are practical and ethical concerns with using these results in the policy process. The greatest concern is that these results become a self-fulfilling prophecy— being identified as at risk of closure could hasten a struggling college’s demise.
Still, would-be students, staff and faculty should do some basic research to protect themselves as they considering enrolling or accepting a job at a college. College employees would also do well to save money and keep their resumes ready; some of these closures are so sudden that employees find out they are out of a job effective immediately and no paycheck is coming next month.
While many data watchers eagerly anticipate the monthly jobs report coming out this Friday, today the Bureau of Labor Statistics released another set of jobs data, and arguably a much better and more complete set of jobs data for 2023. It’s called the Quarterly Census of Employment and Wages, and I have written about this data before.
The QCEW data is better because, as the name implies, it is a census of employment, rather than just a survey, meaning it is an attempt to measure the universe of employment (or at least, the universe of employment covered by unemployment insurance, which is something like 95% of the workforce). Surveys are nice, because they can provide us more timely information — notice that the QCEW is 5-6 months out of date. It is also useful to have this complete data to check on the monthly data and see if it was mostly accurate — indeed, the data is updated through a process called “benchmarking” on a regular basis.
What do the latest QCEW show us? The headline number is that total employment grew by 2.3 million jobs from December 2022 to December 2023, which is 1.5% job growth (if we use annual averages, growth is a little stronger at 2%). That’s a healthy rate of job growth, but it’s less than the familiar Nonfarm Payroll series (CES) shows from December to December: about 3 million jobs added, or a growth rate of 1.8% If we focus just on private-sector employment, we see again that the monthly series is running faster than the more comprehensive QCEW: 2.3 million jobs in the monthly report added versus 1.7 million.
Does all this mean that the monthly jobs numbers are “fake”? Of course not. Surveys will always be imperfect, but they are still useful. But it does mean that you might want to discount them by about 25 percent.
Many people have nostalgia for nominal prices of the past. I’ve written about this topic in various contexts before, but the primary error in doing this is that you must also look at nominal wages from the past. Prices in isolation give us little context of how affordable they were.
One area with a lot of nostalgia is food prices of the past, specifically grocery prices (I’ve also written about fast food prices). While I have addressed grocery price inflation since 2021 in another post (it’s bad, but probably not as bad as social media leads you to believe), there is another version of grocery price nostalgia that goes back even further. For example, this image shows up on social media frequently with nostalgia for 1980 prices:
(Note that the image also mentions housing prices, but the clear focus of the image is on groceries. I won’t dig into housing in this post, but it’s something I have written a lot about before, and I would recommend you start with this post on housing prices from February 2024. But she sure looks happy! As models often do in promotional photos.)
Could you buy all those groceries for $20 in 1980? And how should we think about comparing that to grocery prices today?
One approach to grocery affordability is to look at how much a family spends as a share of their budget on food and other items. In the past I’ve used this approach to show that food spending has fallen dramatically over time as a share of a household’s budget, including since the early 1980s. But perhaps that approach is flawed. Maybe housing has got more expensive, so families are cutting back on food spending to accommodate for that fact, but they are getting less or lower quality food.
For another approach, I will use Average Price Data for grocery items from the BLS CPI series. Note that I am using actual average retail price data, not prices series data, which means there are not adjustments for quality changes or substitutions. No funny stuff, just the raw price data (the only adjustment is if product sizes changes, which of course we want them to do, so we aren’t fooled by shrinkflation — so BLS uses a constant package size, such as 1 pound for many items or a dozen eggs, etc.).
The items I have chosen out of the 150-plus price series are the 24 items which are available in both 1980 and 2024. There may be some biases by doing this, but in general BLS is continuing to collect data on things that people continue buying. So it’s the best apples-to-apples comparison we can do (note that there are no apples in this list! Apples are tracked in the CPI, but there is no continuous price series from 1980 to 2024 for one apple variety).
How best to compare prices over time? Rather than “adjusting for inflation,” as is common in the popular press and by some economists, a better approach that I and other economists use is called “time prices.” Time prices show the number of hours or minutes it would take to purchase the good in two different years, using some measure of wages or income (I will use both average and median wages in this post). By looking at prices compared with wages for individual items, we can see whether each items as well as the entire basket has become more or less affordable.
Here is what time prices for these 24 items look like if we use average wages (I use a series that covers about 80% of the workforce, but excludes supervisors and managers). For this chart, I use prices in April 1980 and April 2024, since there is some seasonality to some prices (and April 2024 is the most recent price and wage data available, so it’s as current as I can get).
The chart shows that for 23 out of the 24 items, it takes fewer minutes of work to buy the items in April 2024 than it did in April 1980. For many items, it is a huge decrease: 13 items decrease by 30 percent or more (30 percent is also the average decrease). And while we once again might be concerned by selection bias of the goods, we have a nice variety here of proteins, grains, baking items, vegetables, fruits, snacks, and drinks. Unfortunately for the bacon lovers out there it is the one product going in the other direction, but there are still a variety of other proteins that have become much more affordable (pork chops are much cheaper!).
Here’s one way in which the image of the lady shopping wasn’t wrong: you could get a basket of groceries for about $20 in 1980. The basket I’ve put together (which is obviously different from the woman’s basket, but you work with the data you have) would cost $27 if you bought the package sizes BLS tracks (e.g., one pound for most of the meats and produce). In 2024, that same basket would cost $84. That’s 3 times as much! But since wages are over 4 times higher, the family is better off and groceries are, in a real sense, more affordable.
Speaking of wages though, is my chart perhaps biased because I’m using the average wage? What if we used another measure, such as the median wage? For that, I can use the EPI’s median wage series (which comes from the CPS), and I also converted it to a nominal wage for 2023. This wage data is only available annually, with the most recent being 2023, so I will also use 2023 price data for this chart (note: for oranges and strawberries, I use the second quarter average price, since they weren’t available year round in 1980 — another subtle example of growing abundance and prosperity today).
The immediate thing you will notice is that there isn’t much difference between the average wage chart. Bacon is still less affordable. We know have oranges being slightly less affordable and strawberries being basically the same, though keep in mind as I mentioned above the chart that these weren’t available year-round in 1980.
But other than bacon and those seasonal fruits, everything is more affordable in 2023 than 1980. The average decrease is the same as the prior chart: 30 percent fewer minutes of work at the median wage to purchase this basket of goods, with 13 of the 24 items decreasing by more than that 30 percent average. The reason for this similarity is that both the average and median wages as measured by these series are more than 4 times higher than 1980.
But are these 24 items representative of other grocery items that we don’t have complete price data in the public BLS series? They are probably pretty close. The unweighted percent change in the items from April 1980 to April 2024 was 201%. If we use the CPI Food at Home component, which includes many more items but also changes in composition as buying habits change, we see a slightly larger 255% increase. But that is still less than wages have increased since 1980 (by over 300% for both average and median wages). As our incomes rise, we will naturally switch to better and more expensive foods, which can explain the 255% vs 201% difference in price increases, but it also shows the BLS isn’t engaging in any funny business with the indexes: if they kept the basket of goods constant, price increases would be smaller.
While the rise in prices since 2021 might rightly make us nostalgic for the pre-pandemic era of prices, let’s not be nostalgic for 1980 grocery prices.
Has the US tax and transfer system reached an egalitarian ideal? That’s one reading of this new working paper “How Progressive is the U.S. Tax System?” by Coleman and Weisbach. After accounting for all taxes and transfers (red lines in the charts), Americans across the income distribution saw roughly 250% real gains in income since 1966:
While market income has grown faster at the top of the income distribution (especially the top 1%), we also tax the rich heavily and use much of that tax revenue to fund direct transfers to poorer Americans and fund programs (such as Medicaid) which benefit poorer Americans. Put it all together, and everyone has seen similar gains over the past five decades, and these gains are fairly large: no Great Stagnation!
Economist Writing Every Day 
