Why Everyone Feels Poor

Americans are richer than ever before. So why does everyone feel poor?

If you spend any time on social media, you will constantly hear that most Americans are less well off than previous generations and getting poorer. However, the statistics show completely the opposite story: every decile of Americans has more real income^[0] than at any time in the past.

I think there are several plausible reasons for this disconnect between the statistics and people's feelings, each of which helps explain a part of it. I will necessarily spend more time looking at technical economic explanations, but I believe these are largely rationalizations, and that the social and psychological considerations I discuss later, while harder to prove, have more explanatory power.

Are We Actually Poorer?

First we should entertain the possibility that, despite what the statistics say, we are in fact poorer in the economic sense than in the past—or at least poorer than the statistics suggest. There are a few plausible lines of argument for this.

Understating Inflation

The most common argument is that CPI drastically understates inflation, for one of the following reasons:

• Technological improvements in consumer electronics, cars, appliances, etc. are over-valued
• The quality of most goods has decreased across the board (food tastes worse, fabric is artificial, etc.)
• "Essentials" such as housing and childcare should be considered massively more important than others
• (far too many much stupider objections to cover here)

But these objections all fail to close the gap. Removing newfangled consumer goods and sticking with used cars and appliances has only a small effect, as these are small components^[1] and the differences in CPI between the new and old items are small. When new, possibly lower quality versions of goods are introduced, the BLS goes through great pains to adjust for their quality^[2], taking into account properties such as the fabric of a suit or whether it is hand-stitched^[3]. The "higher quality" old style is still available for most goods, but it appears consumers generally find the trade-offs of newer goods preferable. The main "essential" goods and services that people complain about are already very large components of CPI. Further upweighting certain components only suffices if you place a massive premium on the most labor-intensive services, such as childcare and cleaning.

And this is ignoring the ways in which CPI may overstate inflation. The most prominent example is that new technologies may make older goods obsolete. For example, CPI includes albums, but nothing in CPI accounts for the shift from having to purchase individual albums to streaming almost every album ever sold for $12 per month. Mainstream economists believe that, on net, factors like these outweigh the factors that cause CPI to understate inflation^[4].

Falling Surplus

A more sophisticated explanation for the gap may be that we underestimated past real income because we only measure market prices, not value—i.e. ignoring consumer surplus—and that consumer surplus has fallen as markets have become more efficient. There are certainly cases where this is true:

• Housing in and travel to beautiful places has become much more expensive relative to other places, so people who already lived there or were in-the-know have been forced to spend more or been priced out.
• A good local restaurant would have been cheaper and less crowded in the days before online reviews, accruing surplus to those lucky enough to live nearby.
• Algorithmic pricing and other price discrimination techniques allow sellers to reduce consumer surplus to levels impossible with fixed pricing.

Yet there are as many or more counter-arguments to this trend:

• The vast majority of people never benefited from these beautiful places in the past, and the many who travel or live there now gain at least some surplus value from them.
• There are many more good restaurants at many price levels now than in the past, with online discovery mechanisms providing people with access to them and improving them through competition.
• As most commodity prices plummet relative to income, consumer surplus from these increases correspondingly.

However, a special case of this is both obviously true and holds significant explanatory power for half the population: increasing opportunity for women has erased a massive amount of surplus value accrued to men due to undervaluing their time and preferences. It was certainly cheaper to have a nice home-cooked meal in the past than it is to eat a nice restaurant meal today—if your wife has nothing else to do. Similarly, keeping a house clean appears on paper much cheaper in the past than hiring a regular housekeeper does today. Some people will argue that we are still poorer, because a typical household cannot afford to have a stay-at-home parent, but a line-by-line comparison of the cost of what was purchased by such households then and now shows it is still perfectly possible—our revealed preference is just to have two breadwinners and more spending power. Those who are incredulous about this are largely men, comparing a past where they would dictate nearly all of the surplus spending decisions to a present where they dictate half (or perhaps less).

There is also the argument that more rational business spending has resulted in less surplus accruing to employees and the public. Perks of upper-class jobs have become stingier, with fewer dinners and less alcohol flowing around the office. Flying business class now is both cheaper and better than flying was pre-deregulation, but employees are increasingly likely to be forced to fly economy. Businesses may be investing less in architecture and public art.

Market Overfitting

A more troubling hypothesis is that, as markets have become more efficient, market value has become more disconnected from "true" value. This is a common critique in Marxist thought, but does not require adopting a labor theory of value, or any specific theory of value. It only requires believing that market value is an approximation of whatever notion of value is most important to humans, even if (as I believe) it is the best approximation available. This can be thought of as an example of overfitting in statistics/ML, or as an example of Goodhart's law.

Market value being our best approximation makes this nearly impossible to prove in principle. However, there is some evidence that at least suggests this could be true. There is the entire field of behavioral economics, which shows that people do not always behave in ways that are compatible with any single notion of value. And we all have anecdotes about our friends and family buying things they will not get value out of and refusing to spend money on the things they will—the introspective will also recognize this in themselves.

I think this is a plausible argument, if only because the phenomenon of overfitting is so ubiquitous in the presence of heavy optimization. But it is unclear how we could even begin to estimate its effect size.

Merely "Feeling" Poor

Since it seems unlikely that we are poorer in the standard economic sense, where does this feeling of "feeling poor" come from?

Relative Wealth

The most obvious criticism of the standard economic story I've seen is that people care more about relative than absolute economic standing. I think this is true and explains away a large portion of the gap. For certain groups which have experienced relative income declines—mainly white men—this might be enough to explain feeling poorer. In a subtler example, the declining value of being in-the-know about where to travel, which restaurants to eat at, etc. disproportionately affects the kind of people who write articles and go viral on social media. However, it is not immediately clear how it could make the majority of people feel poorer.

One possibility is that the set of peers people compare their standing with has changed—in particular, social media disproportionately exposes us to people living expensive lifestyles. Another possibility is that we compare ourselves not to the median peer but to some higher percentile, and as inequality has increased this reference peer's lifestyle has diverged more and more from the median. I believe both of these play a role, but are not sufficient to explain why the perception is so widespread.

Increasing Materialism

Another simple explanation is that, as we have become a more secular and atomized society, people simply care more about material wealth than before. This can help explain people's discontent when combined with another hypothesis: that people mostly view their material wealth through a negative frame (e.g. in terms of their unfulfilled aspirations).

Increasing materialism may itself induce this negative framing. A man who is unsuccessful in love is probably not alone due to his plain car and inability to afford fancy restaurants, and would probably be best served improving his fitness, fashion, charm, and expanding his social circle. But in an increasingly materialistic world, he thinks of his shortcomings primarily in material terms and thus experiences his lack of romantic success as a lack of material wealth.

More Unattainable Goods

Another reason why people may be experiencing their material wealth through an increasingly negative frame is that there are always more goods they can't afford. If you are focused on the goods and services on offer that you can't afford rather than those which you can, then you naturally will feel poorer as more luxury goods and services are invented. In fact, the modern economy delivers a one-two punch by both inventing ever more high-end goods and services and advertising them more ubiquitously.

Decreasing Marginal Utility

If indeed people increasingly frame their discontents in material terms, this suggests a relatively straightforward explanation for why people feel poorer even as they get wealthier: the decreasing marginal utility of money. If we think of "feeling poor" as the gap between the amount of money someone has and the amount they believe they would need in order to feel content in life, then it is obvious that the richer an otherwise discontent person is, the poorer they will feel.

Conclusions

So what can we do about this? The technical explanations suggest some marginal changes we could make to economic policy that might help, but only marginally. Taxes targeting the wealthy, luxury goods and services, and advertising could help reduce the sting of inequality and reduce the visibility of unattainable goods. But the primary issue appears to be a crisis of meaning in people's lives which is being misinterpreted as an economic problem. I am of course not the first to observe this, e.g. Francis Fukuyama^[5]:

The end of history will be a very sad time. The struggle for recognition, the willingness to risk one's life for a purely abstract goal, the worldwide ideological struggle that called forth daring, courage, imagination, and idealism, will be replaced by economic calculation, the endless solving of technical problems, environmental concerns, and the satisfaction of sophisticated consumer demands.

However, we have made no progress on solving this dilemma—if anything, the dire situation Fukuyama foresaw seems all the more certain today.

0. ^

   Remember "real" means "adjusted for inflation". Note I am counting transfers such as Medicaid and SDI here, without which it probably does not hold for the lowest decile but still holds for the rest. Note also that "income" includes both wages and health insurance, which many of the people I'm disagreeing with exclude from compensation but curiously still include (to the benefit of their argument) when calculating inflation.

1. ^

   Component weights for 2024 available here.

2. ^

   See "Item replacement and quality adjustment" and "Direct comparison" in the BLS's handbook for CPI calculation.

3. ^

   See Quality Adjustment in the CPI: Men's Suits. This is not a cherry-picked example; it is the very first entry in the CPI ordered by ELI (AA011).

4. ^

   E.g. this was the conclusion of the Boskin Report, although some but not all of these points have been addressed by improved quality weighting^[6].

5. ^

   From his 1989 essay The End of History? which he later expanded into a book.

6. ^

   See page 10 (as numbered) of this BLS working paper.