Megan McArdle asks an age-old question: what do we mean by “living standards”? The fact that it’s age-old suggests it’s unanswerable, which in itself is interesting. Many of us throw the term around, and one comes away from her essay reminded that we should be clear about what it is we mean when we talk about living standards.
She herself doesn’t answer the question. That’s not a critique—she wisely doesn’t try to. Instead, she lists a few pages worth of ways in which progress—mostly technological, though there’s one particularly important exception I highlight below—has made us better off compared to earlier cohorts. We’ve got better cars, food, appliances, homes, and so on. McArdle recognizes that there’s a distribution of these goods—she’s not saying poor people have homes that are as nice as those of the rich. She’s saying that the homes, cars, computers, smart phones, and entertainment options of today’s poor are a lot better than those of yesterday’s poor.
That’s all surely true. Every generation is better off in this sense than those that came before. As she says, “we should never pooh-pooh economic progress.”
The question is as follows: what do we learn from these observations? McArdle correctly warns that any given statistic, like median wages or GDP, doesn’t tell the whole story. But again, that’s not news, nor should we dismiss such measures as unimportant, simply because they fail to capture all the nuances of living standards.
Her essay thus left me with three questions that I’ll try to answer. First, what reference point do people tend to use when they think about all of this progress? Second, given the context McArdle provides, what is the proper interpretation of important trends like stagnant median earnings or rising inequality? Third, especially since this series is being hosted by the libertarian Cato Institute, it seems important to look at the role played by government in much of the progress McArdle documents.
While I enjoyed and appreciated McArdle’s brief review, she devotes too little space to a critical factor: our reference points. To whom do we compare ourselves when evaluating the progress she documents? Let me explain by way of one example that I think she gets wrong: liberty.
McArdle argues “life is a lot better than it was in 1930 if you’re black” or gay or a woman who wants to work in the paid labor market. Of course, there’s no question that discrimination and violence against such groups is much diminished from the 1930s and even more so—far more so for blacks, many of whom were enslaved—if we were to go back to 1830.
But what contemporary black person (or gay person, woman, etc.) would take solace in this comparison? Their reference point is not 1830 or 1930. It’s not even today. It’s the liberty enjoyed by white people and non-gay people and men today. Under what system of justice is discrimination acceptable relative to the level of discrimination 100, or even 10, years ago? Should we not compare it to the standard of those who do not suffer discrimination today? People do and should compare the extent to which they are “liberated” relative to those who do not face discrimination today.
McArdle is thus wrong to discount this more relevant comparison for liberty. I’m less sure about the proper frame for economic comparisons, but I also suspect that the most commonly applied reference point for today’s poor is not yesterday’s poor but today’s middle class and wealthy. Those living in substandard housing by today’s standards likely take little comfort in the fact that they live in a mansion compared to a century ago.
One bit of evidence in support of my suspicion comes from subjective poverty measures. As economist Rebecca Blank shows here (see her Figure 4), if you ask people what it takes to get by in their community, they reference amounts that track about half the median, leading Blank to conclude “…that people think about economic need in relative rather than absolute terms.”
Thus, whether it’s liberty or household appliances, while we should appreciate that there’s been significant progress over time, our living standards are surely also a function of the resources and liberty we enjoy relative to our present day peers.
An interesting wrinkle here comes from recent economic analysis on productivity growth. In their reductionist way, many economists consider productivity—output per hour—to be a proxy for living standards. And in fact, many of the gains McArdle references show up in the productivity accounts. Computer prices, for example, which are adjusted for all the cool things they can do now that they couldn’t do before, have fallen by half in the national accounts just in the last decade, significantly boosting output and productivity.
But a real concern among economists is that productivity growth has slowed considerably in recent years, implying that while living standards by this metric are still improving, they are doing so at a slower pace.
How does this productivity slowdown relate to McArdle’s piece? Economist Robert Gordon argues that the really important living-standard improvements are in fact all behind us. We would never trade, he argues, plumbing, running water, and air conditioning for Twitter. “Invention since 2000 has centered on entertainment and communication devices that are smaller, smarter, and more capable, but do not fundamentally change labor productivity or the standard of living in the way that electric light, motor cars, or indoor plumbing changed it.” Gordon worries that productivity will continue to grow at historically slow rates, as we’ve already harvested the living-standards game changers.
That conclusion sounds pessimistic to me, but it is true that unless our measurement is seriously biased downward (and I know of no good evidence to support that claim), productivity has grown by 1.3% per year since 2004, compared to 2.5% 1996-2004. Even while I’m with McArdle that productivity and “living standards” are not synonymous, they’re surely intimately related, and historically slow productivity growth potentially bodes ill for her generally upbeat story.
Next, McArdle argues that her evidence of broad progress “…illustrates the poverty of trying to measure living standards by staring at median wages.” How so? Labor income represents about 80 percent of middle-class incomes for working families. The hourly wage is the fundamental building block of that income, and the increase in income inequality means that the median is now a more representative statistic of middle-class household income than is the average.
The fact that, according to the Economic Policy Institute, the real median hourly wage is up only 6% since 1979—0.2% per year—does not of course paint the full picture of living standards. But it does tell you that in order to be able to get ahead, more family members will need to work for more hours per week and more weeks per year (gender breakdowns show that women’s real median wages rose 22% over these years while men’s fell 9%).
Moreover, in a forthcoming paper for the Peterson Foundation, Ben Spielberg and I show solid linkages between inequality of outcomes—stagnant earnings and incomes amidst growing GDP – and the opportunities available to children in families on the wrong side of the inequality divide, including worse educational, health, and, later in life, employment and earnings outcomes. Again, no parent I know would take solace by being reassured that even though these wage problems meant her kid won’t go as far as her capacity could take her, she’ll still do much better than a far more affluent kid from an earlier time period.
Because the stakes are so high in today’s terms, which is the reference point for people’s actually lived experiences, it’s not time to stop staring at stagnating real median wages and the growing gap between worker compensation and productivity. Instead, we must figure out how to reconnect productivity gains with median income growth, and, in so doing, reconnect the economic well-being of large swaths of working families to growth in the overall economy.
Finally, McArdle’s essay raises an interesting question for our host, the Cato Institute: how does an institute that advocates for much less government in our economy think about the fact that government’s fingerprints are all over the advances she celebrates? Al Gore may not have invented the Internet, but like so many of these advances, the key research in its development took place in government (DoD) labs (as Brookings’ Kemal Dervis points out, the government accounted for at least 31% of R&D spending in the US in 2012). The progress in liberty that has occurred, and it has been significant, has crucially depended on federal preemption of local prejudices. Progress against pollution, also cited by McArdle, has largely occurred due to government regulations to “internalize an externality”—to make polluters pay for the price of the damage they inflict on the environment (and, of course, there’s much more to be done here). The increase in travel McArdle cites depends on public infrastructure.
Arguably, if Cato had its way, McArdle and the rest of us might have less to crow about!