One curious detail from the rhetoric of the last two elections was that the political conversation only focused on “the middle class,” as if all sides refused to acknowledge that the poor had any relevance. And since lots of others have bandied the term about, I’m feeling free to jump on the bandwagon here.
The following quote from Foster and Wolfson provides a fairly standard summary of the importance of the middle class:
“The presence of a sizable, well-off middle class is typically presumed to be an important factor in the growth and development of today’s successful industrial economies. The middle class provides much of the labor force for the economy and is a key market for the national product. A large portion of a country’s tax revenue is collected directly or indirectly from the middle class. It also has a special role in the relative political stability these nations have enjoyed. According to Lester Thurow [see their footnote 32,] ‘A healthy middle class is necessary to have a healthy political democracy. A society made up of rich and poor has no mediating group either politically or economically.’” (p. 248)1
I’d like to consider with you some different, economics-oriented ways to define the middle class—because I’m an economist—and then discuss some reasons why the economic health of the middle class seems to have been declining since the early 1970s.
Because there are many, many different components to all of this, I have slowly come to realize that my “unifying theme” will be the following: there are two sides to labor’s impact and importance within the U.S. economy, and both are important and must be considered in all policy decisions.
One side to labor, where the chief focus has been for the last 40-plus years, is as a cost of production.
Table 1 (found in any common Econ 101 textbook) shows the basic framework used in economics, where all the various resources needed to produce anything can be divided into just these four categories.
“Labor’s share” is a very large component in the current debate over the health of the U.S. economy and has historically been close to 80 percent of the nation’s output, when all fringe benefits are added to wages and salaries, and you include the profits of very small businesses, where the owner is providing most of the hours worked. I will return to this point later in this article.
Labor as a cost component comes from looking at any business, making and selling any product (which includes “services,” like getting your teeth checked or advice from a lawyer) and not just tangible “things.” The company focuses on “will it make a profit?” calculated as Total Profits = Total Revenues - Total Costs
And labor’s wages and fringe benefits are a cost, just like the costs to run the machinery. So when sales drop, U.S. managers have been quick to fire workers in order to help lower costs and protect profits. Later we will consider how managers could meet the same goal in a different way that would better help their company, their workers and the economy as a whole.
The second aspect of labor that has been receiving far less attention (in my opinion) is their role in the economy as consumers. The wages and fringe benefits paid to workers—at all levels—are a cost to the company, but at the same time they are the source of income to the worker and her/his family, which plays a vital role in allowing people to be consumers—i.e., customers for businesses, their own and others—so that the company can sell what it makes, and make enough profit doing it, so that they can stay in business going forward.
The fact that real wages, i.e., the purchasing power of wages, has been eroding since the early 1970s (for a variety of reasons) is one big factor in why the current recovery of the U.S. economy has been so anemic.
Let’s begin by reminding ourselves that there is no “economics-only” definition of the middle class. The standard definitions contain a large measure of social elements, which won’t be included here. Economists look at the “personal distribution of income” and the “functional distribution of income,” which give exactly the same dollar figure for any given year, but do the calculation differently.
The personal distribution of income adds all the money coming into each household during the year and then ranks all the households from the richest down to the poorest. Since that would give over 117 million numbers,2 statisticians typically divide the list into groups so the information is more manageable. But the “average income” and the “median income” are calculated from the entire set of incomes for all households.
For the average, you add all the individual income numbers and divide by the total number of households that year, which is also called the mean. The problem with this number is that averages are easily distorted by the very large values among the richest, but relatively rare, households. So economists generally prefer to track the “median income.” To find it, visualize that string of values of income for each single household and go down the list until you find the one that is exactly in the middle, i.e., 50 percent of households are richer and 50 percent are poorer. So we tend to feel that at least this one family should be included in the middle class.
Let’s also keep in mind the difference between “income” and “wealth.” Economists often divide concepts into “flows” and “stocks.” Income is a flow because, like water going through a hose, you need a start time and a stop time to measure how big the flow has been. Income is the “flow” of money to a household, only between Jan. 1 and Dec. 31 of a given year. Wealth, on the other hand, is a “stock” because it can accumulate over time. You can think of wealth as money you have saved from last year’s income and still have, so it is part of your wealth, like “the first dollar I ever earned.” And your wealth can keep growing, year after year, as long as it “makes some money” or you add to it. Income data is fairly easily available, but wealth data is harder to compile and is less often available in the media.
Chart 1, from the U.S. Census Bureau and taken from Wikipedia,3 is unusually valuable because of the level of detail presented.
Each bar shows the percentage of households who are making that much (personal) income, as of 2010. It starts on the left with just $5,000 of income (for all in that household) and rises in $5,000 increments. The first label provided, as you move to the right, shows the median income at about $50,000. So one-half the nation’s households are below that $50,000 income level, yet it is only about 25 percent of the way across all the income increments up to $200,000 of income per year (just before the big spike at the far right), which we might see as 100 percent of the income range for 96 percent of American households. The highest median income in the U.S. had been $55,000 per year, just before the current economic crisis in 2007. Perhaps this graph helps clarify why “47 percent of people in the U.S. aren’t paying any income tax” and why it could be assumed that they don’t need to do so.
The comparison above is of the degree of income inequality in the U.S. in 1970 to the most recent data available, 2010. It takes the “string” of household incomes, from bottom to top, and then divides it into “quintiles,” or five equal parts, so that 20 percent of all the households are in each group (across the lower axis) and shows the percent of income going to all of the households within each group. Two of the more important concepts shown are that: (a) income is quite unequally distributed in the U.S. and (b) the degree of inequality has risen since 1970.
Now, take a timeout in reading this and in the margin write down the top and bottom income levels that you feel should be used to define “the middle class” in America today. Go ahead—remember, there is no “right” answer. And now you can compare your range to the data that follows.
Chart 3 again presents all of the personal distribution of income, but divides it by income levels (on the vertical axis) rather than number of households. The top (darker) bar gives the percent of all households making that level of income (which can include multiple people), while the lower (lighter) bar gives the percentage among single individuals or “persons.” The top category shows that only 17 percent of all households have income of $100,000 per year or more—which includes all of the ultrarich, with millions or billions of dollars of income per year. So might it not be safe to say that any person or household in the U.S. with $100,000 of income per year or more is above middle class, compared to the $250,000 or $400,000 used in modifying the tax rates from the President George W. Bush era?
Part of what makes defining the middle class so complex is the fact that the “purchasing power of money” varies a lot from one state or region to another. You are already aware that Nebraska has a much lower “cost of living” than does New York, Connecticut or California, for example. What you can buy for $50,000 when you focus on food, rent, transportation, etc., varies widely, and yet the U.S. has a long tradition of using just one number for many federal policies provisions, unadjusted for cost of living, because it is considered more “fair.”
Charts 4 and 5 each come from Wikipedia and present different aspects of who are the “winners and losers” in the current American distribution of income.6 The figures given are the average of personal income per year for all the individuals within each group. Ethnicity and gender are powerful factors in our income distribution. We each carry within ourselves a lot of reactions to this largely “social” information, but we need a different economist, specializing in labor, to provide further economic analysis.
Per capita income is calculated as a ratio. The total of all personal income for a given year is in the numerator, and it is divided by the total population. This number is most useful for doing comparisons over time or across nations. Chart 6 provides that information for the period 2001 to 2011. But please don’t read it as showing large changes because the left-side scale has only $500 difference between each of the levels of income presented—but it does show you the trend.
So how might we define the middle class using just income levels? The easiest way is to use statistics and focus on a range on each side of the median income. One typical suggestion is to use between 75 percent and 125 percent of the median income.7 For the current median income of $50,000, that would mean that the middle class starts with families earning $37,500 and ends with those earning $62,500. My own entirely unscientific view is that those numbers are too low, given the stagnation of real wages in this country since the early 1970s, and that it would be OK to use $50,000 up to $75,000. What do you like about your range?
One reason I like $75,000 in household income as the cutoff between middle class and the rich is that research by Daniel Kahneman indicates that, in general, “happiness” rises as income rises, but only up to about $75,000.8 After that, even with more and more income, happiness stays at the same level! In Chart 2 the internal labels indicate that only 25 percent of households reported incomes of over $85,000, so that could also be the top cutoff for the middle class. And since only 10 percent of households reported incomes over $135,000 it seems to me that, for a nation as wealthy as the U.S., those in the top 10 percent have to accept the label “rich.”
Now let’s switch to another facet of this topic and discuss further the “functional distribution of income” introduced in Table 1. This concept is also found in the “National Income and Product Accounts” of federal data, but the definitions are complex, and detailed calculations from the basic components are needed to isolate “labor’s share,” so I am relying here on calculations done by other analysts in their published reports.
Economists are fond of discussing the “capital/labor ratio,” which puts the income shares of all the other three resources (land, capital and the entrepreneur) together and calls it “capital” for the numerator, and compares it to “labor’s share” in the denominator. This is a big-picture concept and has been used to provide the following generic explanation for workers around the world doing less well. Since at least the early 1990s, China and India have become increasingly important players in the world’s economy. Since both of them have an abundance of labor (being added to the denominator for the world) and relatively little amounts of “capital”—think machines, factories, etc.—to add to the numerator, the value of the ratio as a whole has shrunk. Workers face a lot more competition from other workers around the world, so their pay has failed to rise as fast as output has, and the “returns to capital”—the money in the hands of owners of the other resources—has risen, relatively, and workers are getting a “smaller slice of the total pie.”
A more narrow, technical definition of labor’s share of income is this: “Household income comes in two types: labor income, which includes wages, salaries, and other work-related compensation (such as pension and insurance benefits and incentive-based compensation), and capital income, which includes interest, dividends, and other realized investment returns (such as capital gains). During the last three decades, labor’s share of total income has declined in favor of capital income…” (p. 1, Jacobson and Occhino)9
Chart 7 provides a sense of the pattern in the decline in the fortunes of workers, which I am using as a proxy for “the economic health of the middle class.”10
This chart shows the ratio of wages and salaries, compared to the GDP (the total output of the nation, in the same year), and both have been adjusted to take out the impact of inflation. The overall, downward trend, since about 1970, helps show clearly why so many Americans are struggling in today’s economy, even though financial numbers for the wealthiest are growing well.
Recent research on the reasons for this decline also comes from the Cleveland Fed. “Labor income has declined as a share of total income earned in the United States. This decline was caused by several factors, including a change in the technology used to produce goods and services, increased globalization and trade openness, and developments in labor market institutions and policies.”11
Inflation is the final element I’d like you to consider with me.12
Even when a worker has been employed since 1970 at a salary that was a bit above the median income, raises have not tended to keep up with the overall level of inflation. Chart 8 shows the startling—to me—impact of inflation in the U.S., as measured by the CPI, since 1960.12 Serious inflation began in the 1970s with the two OPEC oil crunches (about 1973–74 and 1978–79), and real wages and benefits (the bottom two lines)—i.e., the purchasing power of the salary—have stayed virtually flat from about 1970 until today. (Although the graph stops in the mid-2000s, updates look about the same). So even though the amount on the annual paycheck went from under $10,000 in 1960, and yearly raises about the size of inflation were given, the almost $70,000 paycheck today doesn’t buy any more for this worker. And these workers were lucky—they did get raises that kept up with inflation—but many more of America’s workers didn’t get many raises, and their “bottom two lines” have actually declined since 1970.
In conclusion, it is my opinion that the U.S. economy will continue to grow only very slowly, with the possibility of frequent stagnation, unless the economic well-being and size of the middle class expands considerably to help maintain the consumer as a driving force in the economy.
1. James E. Foster and Michael C. Wolfson, Introduction to “Polarization and the Decline of the Middle Class: Canada and the U.S.,” “Journal of Economic Inequality,” vol. 8 (2010), pp. 247-273; Lester C. Thurow, “The Disappearance of the Middle Class,” New York Times, F.3, February 1984.
4. Chart 2 was compiled by the author from Table A-3, Selected Measures of Household Income Dispersion 1967 to 2010, www.census.gov.
8. Daniel Kahneman is a noted psychologist and won the Nobel Prize in economics for work that was foundational in the creation of the relatively new “behavioral economics,” which looks at how people actually make decisions, quite apart from the assumption of “rational behavior,” which has been assumed in neoclassical economics from very early days. References to his $75,000 threshold in his research on happiness have been widely quoted, as in Jean Chatzky’s “Top 5 Money Rules for 2013”: “5. More Money Won’t Always Make You More Happy,” www.marketplace.org/topics/your-money/personal-finance-reference-guide/top-5-money-rules-2013.
9. Margaret Jacobson and Filippo Occhino, “Labor’s Declining Share of Income and Rising Inequality,” “Economic Commentary,” Sept. 25, 2012 (Cleveland Fed), www.clevelandfed.org/research/commentary/2012/2012-13.cfm.
12. Chart 8 is from Nebraska data in the author’s archives that have been generalized to not reveal the particular job category. The data accurately show the impact of inflation on both wages and benefits, for the same workers year by year.