Subject: Explaining Growing Wage Inequality Date: Mon, 14 Dec 1998 14:35:30 -0500 From: Edward Flaherty Organization: College of Charleston Newsgroups: sci.econ, alt.politics.economics Since about 1980 income inequality has risen in the U.S. by every statistical measure. In this modest essay I present several measures of income inequality over the last 18 years and then offer several explanations for the growing income gap. First, let's take a look at some facts about income inequality. Presented below are four measures, all of which account for inflation. None, however, incorporate worker benefits except the last measure. For example, in 1994 the three lowest quintiles of families received 29.9% of aggregate income, while the top two quintiles received 70.1%. In 1980 those respective figures were 34.2% and 65.9%. See the U.S. Census Bureau website at http://www.census.gov/hhes/income/incineq/p60tb1.htm The mean income of the three lowest quintiles of households also fell during that period, while the income of the top two quintiles increased. See http://www.census.gov/hhes/income/incineq/p60tb3.htm One measurement problem with these figures is that they say nothing about mobility between the groups. It is not true that it is the same people every year enjoying the top quintile and the same mired in the bottom quintile. A worker in the top quintile this year is probably enjoying the best income year of his life and is statistically likely to drop down closer to the mean in following years. Likewise for those currently in the bottom. They will likely rise closer to the medium in future. According to Brooks Pierce, economist at the Dept. of Labor, the wage gap between highly educated workers (those with at least a bachelor's degree) and less educated workers (those with a high school diploma or less) widened considerably. In 1982 high-ed workers made $24.80 an hour, 3.92 times the $6.28 made by low-ed workers. In 1997 high-ed workers received $25.74 an hour, 4.72 times the $5.46 an hour that low-ed workers received. So real wages for the highly-educated increased while the real wages for the lesser-educated fell. But wages are only one form of worker compensation. The above figures do not take into account the value of fringe benefits, both voluntary and mandated, that have become a large part of total employee compensation. If we include benefits, then high-ed workers made $35.16 an hour in 1982, 4.56 times the $7.72 figure for low-ed workers. Note that in 1982 including worker benefits narrowed the total wage gap between the groups. In 1997, however, high-ed workers made $36.89, 5.43 times the $6.79 for low-ed workers. Factoring in benefits actually worsened the inequality in 1997. Another set of facts which will be useful it the share of total employment going to high-ed workers and low-ed workers. Men with at least four years of college saw their share of total U.S. employment rise from 11.4% in 1980 to 17.7% in 1992. The figures for women with a four-year college degree are 10% in 1980 and 17.2% in 1992. Men with only a high school diploma saw their employment share from from 38.2% in 1980 to 33.6% in 1992. Women saw their share drop from 46.4% to 36.8%. In general, then, high-ed workers saw their employment opportunities grow, while low-ed workers saw theirs shrink. Why the growing inequality? Labor economics suggests three possibilities. The first is a supply effect. The observed increase in wages for high-ed workers could have been due to a decline in the supply of such workers. Given a constant demand curve, then their wages would have to rise. However, if this explanation were correct, then we would also expect to see a decline in the employment of this group. In fact, we observe the opposite. This explanation cannot be a correct one. Similarly, the decline in wages for low-ed workers could have been due to an increase in the supply of low-ed workers because, say, of an increase in immigration. However, if this were true then we would expect to see an increase in their employment. Again, we observe the opposite, so this cannot be a correct explanation. Another possibility is the decline in unionism. Low-ed workers are unionized much more than high-ed workers. Unionized workers earn about a 10-15% wage premium over similar non-unionized workers. The decline in union participation and strength may explain the growing wage gap. There are problems with this explanation, however. Unionism began to decline in the early 1950s, well before the wage gap began to widen in the early 1980s. More revealing however is that the wage gap between high-ed women and low-ed women has not grown nearly as much as for men, yet women are unionized at a far lower rate than men. If the union explanation were an important one, then we would expect to see the wage gap for women growing faster than the one for men. In fact, we see the opposite. A third explanation is demand. An increase in the demand for high-ed workers, given supply, would increase their wages and their employment. This is consistent with what we observe. A decrease in demand for low-ed workers would decrease their wages and their employment. This is also consistent with reality. Therefore, the best theoretical explanation for the growing wage gap is a decline in the demand for low-ed workers relative to high-ed workers. The next logical question is: What caused the changes in demand? One answer is a change in product demand. The demand for labor is a derived demand in that it ultimately comes from the demand for the product labor is helping to produce. Since the mid-seventies consumers have increase the share of their spending going to service-related goods such as health care, and other goods requiring highly-educated labor to produce. They have spent proportionally less on manufactured goods, products which can generally be produced with low-ed workers. Manufacturing and trade, which employ mainly low-ed workers, have seen their total employment share drop from 44.8% in 1980 to 37.7% in 1992. Private services (business, health, legal, recreational) have seen their share rise from 20.6% to 26.8%. (See the monthly 'Employment and Earnings' report from the Dept of Labor). Another significant factor affecting the relative demand for high-ed and low-ed workers is the change in production technology. The early eighties saw the advent of the personal computer revolution as well as substantial moves toward automation and computerized office technologies. These changes affected high-ed workers and low-ed workers differently. High-ed workers and low-ed workers are substitutes in the production process. The improvement in technology made high-ed workers much more productive relative to low-ed workers, and hence made high-ed workers more valuable relative to low-ed workers. Firms substituted high-ed workers for low-ed workers. This can be seen by looking at intra-industry employment figures. In the manufacturing sector the share of total employment going to high-ed workers rose from 17.7% in 1980 to 21.3% in 1992. The share of manufacturing jobs going to low-ed workers fell from 43.5% in 1980 to 41.4% in 1992. In the service industries previously mentioned, high-ed workers saw their share rise from 41.0% to 43.1%, while low-ed workers saw theirs fall from 27.0% to 25.4%. The change in technology increased the demand for high-ed workers (indicating the two are complementary) but decreased the demand for low-ed workers (indicated the two are substitutes). Other explanations for the growing wage gap are outside the realm of labor economics. One is the reduction in welfare benefits during the eighties and nineties. This explanation is a very minor one, in my opinion. Although I lack data on how much income for various quintiles is from welfare programs, the reduction in benefits probably only affected the lowest quintile, or perhaps the two lowest. Yet, the data indicate that the share of income going to the third quintile also fell. The reduction in welfare benefits can't explain the decline for the third quintile. Another non-labor explanation is the Reagan-era tax cuts that reduced marginal tax rates disproportionally in favor of upper-income groups. This explanation also seems to hold little water because tax studies indicate that the overall degree of progressiveness in the U.S. tax system is about the same today as it was prior to the Reagan tax cuts. The labor-related explanations for growing wage inequality are far more compelling in my opinion. They are highly consistent with the data and explain the twenty year trend we continue to observe. This is an important realization because it sheds bright light on ways that will and will not be effective in halting the inequality. More than that, however, is that it raises the question on whether reducing inequality is a policy we should even pursue. -- Edward Flaherty School of Business & Economics College of Charleston Office: (843) 953-7166 Fax: (843) 953-5697 Web site: http://www.cofc.edu/~flaherty REPLYS: From: av139@lafn.org In article <36756881.C3018442@cofc.edu>, Edward Flaherty wrote: > Since about 1980 income inequality has risen in the U.S. > by every statistical measure. In this modest essay I present > several measures of income inequality over the last 18 years > and then offer several explanations for the growing income > gap. The figures I have seen show that wage inequality in the US was on the decrease until about 1970 and has been on the increase ever since. > Similarly, the decline in wages for low-ed workers > could have been due to an increase in the supply of low-ed > workers because, say, of an increase in immigration. However, > if this were true then we would expect to see an increase > in their employment. Again, we observe the opposite, so > this cannot be a correct explanation. This is indeed the correct explanation. About 9% of US workers are foreign born. But yet, at the lowest decile (or quintile) well above 9% of US workers are foreign born. Immigration has been expanding the supply of labor at the bottom at a faster rate than the total labor supply. As late as the early 80's, the meat packing industry paid wages that were middle class and the workers were US citizens. Today, the wages in this industry are near minimum wage, with the US workers being replaced by foreign born -- primarily illegals. This wage depression is also occurring in the professions. From historical statistics I have seen, the engineering profession had a slightly higher salary than the legal profession throughout the first half of this century. Today, engineering salaries have declined to below 70% of the salaries of lawyers. What's going on here? If you go back to the late 60s, both the engineering and the legal profession had no more than a few percent foreign born employed in them. As of the early 90s, the legal profession still employs only about 3.5% foreign born, while engineering employs about 20% overall and at the PhD level 40% Since the mid 60s, employers have had a restricted privilege to recruit foreign born technical professionals. > A third explanation is demand. An increase in the demand > for high-ed workers, given supply, would increase their > wages and their employment. This is consistent with what > we observe. A decrease in demand for low-ed workers > would decrease their wages and their employment. This > is also consistent with reality. > Therefore, the best theoretical explanation for the growing > wage gap is a decline in the demand for low-ed workers > relative to high-ed workers. But these changes in demand are occurring in other developed countries as well without increases in wage inequality. Look at Japan. > This is an important realization > because it sheds bright light on ways that will and will not > be effective in halting the inequality. The way to decrease wage inequality in the US is to filter immigration -- let in only educated immigrants. > More than that, > however, is that it raises the question on whether reducing > inequality is a policy we should even pursue. It better be a policy we pursue. Do you think the massive flow of uneducated immigrants have any understanding of the law of comparative advantage -- let alone market economics? Forget about free trade when they began to dominate the body politics. 3rd world economics here we come -- with its protectionism and crony capitalism that leave such countries basket cases. -----------== Posted via Deja News, The Discussion Network ==---------- http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own AND: Edward Flaherty wrote: > > Since about 1980 income inequality has risen in the U.S. > by every statistical measure. Hi, An interesting post. I have some questions and some comments. First, it is clear to me that there is not one explanation, but that several factors contribute. I think the reply from av139@lafn.org demonstrates that a part of the reason for the lower wages not rising is because of immigration. Of course the immigrants are better off here, and they lower both average wages and prices. > One measurement problem with these figures is that > they say nothing about mobility between the groups. > It is not true that it is the same people every year > enjoying the top quintile and the same mired in the > bottom quintile. Consider income mobility, and the fact that individual incomes tend to rise with age (until retirement). This means it would be possible for the average wage to remain constant, or even fall, at the same time that EVERY WORKER was getting a wage increase every year (even in "constant dollars"). >Therefore, the best theoretical explanation for the growing > wage gap is a decline in the demand for low-ed workers > relative to high-ed workers. While I think this is obviously the main factor, can either of us be "unbiased" on this? I am employed at the UW, you are currently teaching in a college, and the obvious implication of this is that people need more education (ie teachers). Are you promoting your self interest? > Other explanations for the growing wage gap are outside > the realm of labor economics. One is the reduction in > welfare benefits during the eighties and nineties. But your data is on wages, not welfare benefits. I have read in other threads that most welfare benefits (including food stamps and EITC) do not even count in the statistics on wages, or even incomes. > Another non-labor explanation is the Reagan-era tax cuts that > reduced marginal tax rates disproportionally in favor > of upper-income groups. Same for taxes. Are the "wages" figures for "after tax wages"? And what about the claims that the actual fraction of the taxes paid by the top 1% and top 10% of incomes was higher (not lower) during the years of the 28% top bracket than either before or since? -- ,,,,,,, _______________ooo___(_O O_)___ooo_______________ (_) jim blair (jeblair@facstaff.wisc.edu) For a good time call http://www.geocities.com/capitolhill/4834