Imperfect Competition

As we saw in our analysis of output markets, prices can be influenced by market structure. As we moved from a competitive environment toward monopoly, we found that prices tended to rise and output fell. The same is true in the labor market. If the demanders of labor gain market power you will find that employment and wages will be lower than they would be in a competitive market, while if suppliers have control we will end up with lower employment and higher wages. We can also expect that a firm's market power in the output market may affect the earnings of workers. If industry profits tend to be high, which they would be if the firm were exercising market power, then you might find that the workers are better compensated than similarly skilled workers in more competitive industries. This would help us explain why secretaries working for investment bankers earn more than secretaries working for auto dealers.

An additional source of imperfect competition would be labor unions which appeared on the labor market scene in the late 19th century as a reaction to the enormous market power wielded by the newly emerging industrial giants. Attempts to organize labor had been around at least since the guilds of the Middle Ages in Europe, but in the US organization of labor came only slowly. In fact until the Massachusetts Supreme Court acknowledged the rights of unions to exist in 1842, unions were illegal. After the Civil War there were three developments that altered the balance between workers and owners. The first was the massive increase in the scale of operation that increased the separation between owner and worker. Second we had the massive movement off the farms and into the cities.   Individuals were being pushed by the mechanization that freed labor on the farms and pulled by the opportunity of jobs in the new factories in the rapidly growing cities. Third, this was a time of increasing immigration that entered through Ellis Island in NYC and filtered into the cities in the industrial Northeast and North Central regions. In 1880 the nation's labor force stood at 17 million and in the next decade immigration averaged 500,000 a year. If immigration on the same order of magnitude took place at the end of the twentieth century, immigration would average approximately 4 million people a year - more than the number living in Los Angeles. This ample supply of workers from abroad and from our farms moved into the cities and helped keep wages down and working conditions deplorable by today's standards.

One of the results was increased conflict between workers and management that resulted in violence and bloodshed.  In 1872 builders in NYC went on strike and by the following year miners in the coal fields of Pennsylvania had organized a secret society called the Molly Maguires.  The pace and severity of conflict picked up in the following decades with the Haymarket Massacre in Chicago in 1886, the Homestead Steel Strike in Pennsylvania in 1892, and the Pullman Strike of 1894. This certainly helped the side of the owners in the battle for public opinion and government support.

A second result was the emergence of efforts to organize labor. There had been a number of attempts at organization of labor in the aftermath of the Civil War.   We will pick up the story in 1869 as the garment cutters in Philadelphia organized the Knights of Labor which had a goal of organizing all workers into one union and help redistribute income to the workers. This movement died out by the late 1880s, but there was a second current of labor organization underway. In 1881 Samuel Gompers established the Federation of Organized Trades and Labor Unions. The idea was to create an umbrella organization for the nation's craft unions. In 1886 there was a reorganization and the American Federation of Labor (AFL) was born. Under Gomper's leadership the AFL focused its attentions on working conditions and left solution of society's greater problems to others. The unions were organized around individual crafts and one of the primary bargaining tools was the strike. Growth was slow yet steady and by 1904 there were 1.6 million workers out of a workforce of 30 million enlisted in the cause.

Problems were already surfacing, however, that would limit the growth and success of the AFL. Serious problems emerged over the proper boundaries between the separate unions while massive immigration and the spread of large scale factories allowed firms to substitute less skilled, cheaper labor. The depression of 1920-21 and some unsuccessful strikes greatly weakened the AFL which saw its membership slide from 4 million to 2.2 million in 1929. In the 1930s during the Great Depression the movement to organize industrial workers began in earnest and by the end of the decade we had the Congress of Industrial Organization (CIO) that organized workers by industry rather than occupation. The result was that unionization rebounded and by the outbreak of WWII, union membership in the US had risen 150 percent from pre Depression levels.

The growth that took place would not have happened, however, if it had not been for government involvement and changing public sentiment. During the Depression public sentiment turned toward labor's plight and the government altered the balance of power between management and labor with passage of the Norris-La Guardia Act of 1932.  This law greatly restricted the use of the injunctions that effectively defused strikes and yellow-dog contracts requiring workers to disassociate themselves from unions were deemed unenforceable. By the end of the decade we had passage of the National Labor Relations Act, known as the Wagner Act.  This law was based on the premise that the existing balance of power between workers and business was injurious to the workers and the nation. To address the imbalance, the act established collective bargaining as the cornerstone of industrial relations in the US and management was compelled to recognize workers' right to organize. It also set up a National Labor Relations Board to enforce the law.

The power of organized labor was reigned in, however, in 1947 in the aftermath of WW II and shortly after the Republicans took control of Congress in 1946. Over the veto of Harry Truman, Congress passed the Labor-Management Relations Act (Taft-Hartley Act) that limited the power of unions by outlawing strikes by government workers and secondary boycotts where unions could set up picket lines to suppliers of the company being stuck. One of the outgrowths of this was the right-to-work legislation that made it illegal to enforce union-shop agreements that required workers to join unions.

Despite the limitations imposed by the Taft-Hartley Act, unionization of the workforce continued to grow and by its peak in the mid 1950s, more than one quarter of the nation's workforce was unionized. Since then, however, unionization has declined in importance and by the late 1990s unions' share of the workforce was at half its peak level. And this actually understates the decline in unions because the decline has been softened by the growth in public sector unions that by the mid 1990s accounted for more than 40 percent of union membership. The biggest union in this group would be the American Federation of State, County, and Municipal Employees (AFSCM).

There is no single, simple explanation for this decline of unions, but there are a number of contributing factors.  One has certainly been the growth of wages since WW II, at least through the mid 1970s.  Between 1950 and 1973 the average, inflation-adjusted earnings of workers in the US rose approximately 50 percent.

A second has been the relative decline of the manufacturing sector and the rise of the service sector which has been traditionally less unionized because of the smaller scale of operations.

The rapid growth of women in the workforce may have also contributed to the decline since women have traditionally been less supportive of unions.

As for the impact of unions on the labor market, it is not easily captured in the S-D model because there is no single goal that all unions pursue.   For example, a union could attempt to maximize the union wage, the size of the union membership, or the income of the union members. We can, however, make some generalizations based on our knowledge of supply and demand. We could expect union activity to generally fall in the categories of restricting the supply of labor, increasing demand for labor, and reducing the elasticity of demand, all of  which may be accomplished directly in the labor market or indirectly in the output market. For example, union support for import quotas, buy union advertising, and the fight against NAFTA can all be seen as attempts to increase demand and reduce elasticity of demand for union made output. Union support for restrictions on immigration and the minimum wage would be examples of efforts to restrict supply and decrease the availability of substitute inputs. The net result would be an increase in union wages which is where we might look for an explanation of the earnings differences between highly unionized miners and retail workers who tend not to be unionized.

Competitive balance can also be affected by government intervention.  One such intervention would be the imposition of a minimum wage which we looked at in the supply-demand unit.  The market equilibrium would have been reached at the wage where S = D, but the government imposed a price floor so that the equilibrium occurs at the price floor where the amount supplied exceeds the amount demanded resulting in a surplus (unemployment). 

The government could also intervene by restricting the supply which appears as an inward shift in the supply curve.  Examples of these restrictions would be the licensing of cabs and many of the professions where the government establishes entry barriers.  This will tend to keep wages in these sectors above competitive rates and therefore wage differentials would be expected to exceed what could be expected based on compensating differential. 

While imperfect competition may play a role in explaining some earnings differentials, in the discussion of faculty pay differences it is likely to be a non issue if all faculty have the same union status.  This would be the case if the study were conducted for RIU, but if this were a national study you would most likely want to include union status as a potential determinant of earnings. 

Now we have reached the moment of truth.  Do the factors that have been discussed to this point explain all of the earnings differences, or are there some differences left over.  Are there some differences that remain to be explained by factors such as gender and race.  For a discussion of this factor, we will now turn to the section on discrimination.