Poverty, Income Distribution, and Wage Inequality

Broadly speaking, economists define average material standards of living by the Gross Domestic Product per capita. Each person in the United States produces about $25,000 per year in goods and services. And this figure, except for recession years, is increasing. On average then, material living standards in the United States are rising. It is true that they have been rising at a much slower rate since 1973, but they are rising nonetheless. This means that the average person is able to enjoy more goods and services than before. There is no reason to believe that this trend will reverse in the near future.

We always hear the cliche that "this generation will be the first that is not as well off as our parents' generation." Because of the constant march of technology, this statement is simply not true. It is true, however, that the current generation may only be slightly better off than their parent's generation, a result of the slowing of the growth in GDP per capita.

However, statistical averages mask the diversity that exists withing a given cohort. Most of us are not "average," we are either richer or poorer than the average. Therefore, it is informative to look at how our income and wealth are distributed throughout the population. As an example, Kuwait is often cited as having one of the highest GDP per capita figures in the world, however its income distribution is so skewed towards the top that most people in Kuwait are fairly poor. Many of the other oil-producing nations are in the same situation.

Income distribution has direct consequences on the poverty rate in a country. Even in a wealthy country, poverty is a problem if a significant fraction of the poorest people have little or no income. We will examine poverty in the United States and then turn to the issue of income distribution.

THE ECONOMICS OF POVERTY

The official poverty line (click for official 1996 poverty threshholds from the Census Bureau) is the annual cost of a nutritionally adequate diet multiplied by three. If money income falls below the poverty line, people are counted as poor.

Click here to view a Census Bureau table showing poverty rates for families.

Click here to view a chart of the poverty rate.

Money income includes earnings before taxes plus private and government cash transfers, such as alimony and child support payments, Social Security, unemployment checks, and AFDC payments. In-kind transfers (payments-in-kind) are not counted in official poverty estimates. These exclude spending on medical care, food stamps, and housing assistance. Inclusion of these reduces the poverty rate by 2.7 percent, or 6.8 million people. The chart below shows the poverty trend in the United States from 1959 to 1995. As you can see, the poverty rate decreased substantially with the introduction of many anti-poverty programs in the 1960s. But the rates bottomed out in 1974 and have been slowly rising since with little trend in recent years.

The pretransfer poverty rate measures the estimated poverty in the absence of all government cash transfers. In other words, this statistic gives us an indication as to the "core" level of poverty that would exist without any government intervention. This rate increased 3 percentage points between 1975 and 1992, implying that the poverty in the Unites States has increased over the last 17 years.

Check out these poverty statistics published by the Census Bureau.

The Feminization of Poverty

The highest incidence of poverty occurs among female-headed households, blacks, and hispanics. In 1959, 25 percent of the poor were black and 26 percent lived in families headed by a woman. In 1992 the respective figures were 29 percent and 37 percent. Thus the largest increase in poverty has come from female headed households. This has led to what is called the feminization of poverty. The percentage of families headed by women with children who had never married increased from 25 percent in 1979 to 41 percent in 1992.

The Main Government Transfer Programs

Government entitlements are the fastest growing portions of the federal budget. The largest of these programs are Social Security (which is not a "pure" transfer program), Medicare (health care for the elderly), Medicaid (health care for the poor), Food Stamps, SSI (a branch of Social Security for the poor elderly and disabled), AFDC (Aid to Families with Dependent Children), and Housing Assistance. AFDC is often what we mean when we refer to "welfare. As you can see from the chart below, entitlements account for more than half of the federal government's outlays. Social Security and health care alone account for over 40 percent of the federal budget. When traditional welfare programs are included, the number rises to above 50 percent. Entitlements have become huge factors in our economy.

Most welfare programs are income-tested, or means-tested. This means that the amount of the transfer tapers off as the recipient's net income rises. Qualification for programs and the amount of assistance received depends upon one's income. The transfer reduction rate (TRR) is the amount by which the transfer is reduced for every $1 increase in income. For example, if the TRR = 0.3, a $100 increase in income reduces the transfer by $30. All of the above programs are means-tested except Medicaid. AFDC has a TRR of 1 implying that a $1 increase in income reduces AFDC by $1.

Impact of Transfers on Work Effort and Family Composition

Many people criticize current welfare policy because it reduces work effort and encourages the break-up of families. An income-tested transfer reduces work effort in three ways:

  1. The transfer gives the individual more income which is used to take more leisure time.
  2. The individual finds that higher earnings from work cause the transfer to fall due to the effect of the transfer reduction rate. Therefore, work hours are scaled back or eliminated.
  3. If the amount of the no-work transfer is large enough, the individual may decide not to work at all.

AFDC provides a great disincentive to work because it has a TRR of 1, and medicaid benefits are forfeited when one goes off AFDC. The evidence indicates that income-tested transfers cause welfare recipients to work about 5 fewer hours per week. This increases the poverty rate by 1.2 percentage points, or 3 million people.

Transfer programs also contribute to the break up of the family. This is because:

  1. In many states AFDC payments are denied to households where both parents are present.
  2. Public assistance provides unhappy couples the means to establish separate households.
  3. Public assistance increases out-of-wedlock childbearing.

Evidence indicates that public assistance is not a major cause of the growth of female-headed families. The percentage of female-headed families grew in the 1970s and 1980s despite a 42 percent drop in the real value of public assistance. One study estimates that public assistance accounted for only 5 percent to 14 percent of the growth of female headed households from 1960 to 1975.

Policies to Reduce Poverty

It is unrealistic to think that we can eliminate poverty. Poverty, however, is a moving target. As all standards of living change, our definition of poverty also changes. For example in the United States many people living in poverty have running water, electricity, indoor plumbing, decent shelter, food, and perhaps even a car. Poor people in the US in the 1930s had none of these things. Indeed, many wealthy people from other countries don't have some conveniences that even our poorest people have. I was recently in India, staying with a very well-to-do family who had no running water from 9am to 5pm each work day.

Even given this, we can certanly improve and/or revamp our welfare programs to reduce proverty. Some common suggestions are these:

  • Medical protection for the working poor. Health care for all takes away the incentive to go on AFDC to qualify for Medicaid and would all Americans a minimum level of health care. However, this is expensive.
  • Raising the minimum wage. This will help many teen-agers and young workers, not as many heads of households. Nevertheless, it will help to ease the income constraint on many families.
  • Targeted wage subsidies. These are government subsidy payments that supplement the employee's wage. This would have no adverse effect on unemployment but it would be politically unpopular because the burden falls on tax payers, not employers.
  • Expanding the earned-income-tax credit. Families are given credits against tax liabilities. In fact, Clinton has significantly expanded this credit in the last two years.
  • Stepped up efforts to collect child-support. This will ease the financial burden that many single parent households feel.

Income Inequality

Poverty and income inequality are closely related. The more unequal the distribution of income, the higher the incidence of poverty. We measure income inequality by the Lorenz Curve. This is a chart that plots percentage of the households ranked by income on the horizontal axis (ranked from lowest income earner to highest income earner) and the percentage of total income earned on the vertical axis. Two points are always certain. They are the origin and all income. If there are zero people, they earn zero income, and all the people earn all the income. Therefore, the Southwest and Northeast points on the box do not change. The measure of income distribution is represented by the shape of the line that connects these two endpoints. Notice that if the poorest 20 percent of the population earn 20 percent of the nation's income, 50 percent of the population earn 50 percent of the income, and so on, then the Lorenz Curve is a 45 degree line. So the 45 degree line in the chart respresents the line of perfect income equality. But if 20 percent of the population earns only 5 percent of the nation's income, and 50 percent of the population earns only 25 percent of the income, then the Lorenz Curve bows downward. The curve plotted for the United States is for the year 1995.

Another way to express the level of income inequality is by calculating the Gini ratio, or theGini coefficient. This index is simply a meaure of the area between the perfect equality line and the actual distribution line in the Lorenz curve. If income distribution is perfectly equal, the Gini coefficient is zero. If income distribution is perfectly unequal such that one household owns all the nation's income, the Gini coefficient is one. For 1994 the Gini coefficient was 0.456. Since 1967, the Gini coefficient has been rising, again reflecting the increasing income inequality in the United States.

Click here to examine the historical tables of income distribution in the United States as compiled by the Census Bureau.

Explanations for the Rising Income Inequality

There is wide disagreement over what has caused the trend towards rising income inequality. Three possiblilities are presented below.

  1. Increase in wage inequality. When wages fail to keep up, so do incomes. We have more working poor today and the wages of those at the bottom of the income scale have fallen considerably in the 1980s.
  2. Tax and Spending Policies. It is possible that our tax policies are not as progressive as they once were. We are not distributing as much purchasing power to the poor in our society.
  3. Increase in single parent households. Single parent households do not earn as much income as two parent households. This may play a large part in explaining the rise in low income households.

Wage Inequality

The table below presents data on wage changes between 1950 and 1990. The data come from the US Bureau of the Census, and from Juhn and Murphy, Economic Policy Review, Jan. 1995.

                       REAL      WAGE        GROWTH                 

Percentile 1940-50   1950-60    1960-70     1970-80     1980-90    

11-20         .315      .278       .192        -0.15      -.169    

21-40         .277      .292       .207        .015       -.116    

41-60         .197      .301       .232        .073       -.072    

61-80         .127      .302       .252        .096       -.024    

81-90         .091      .300       .284        .089        .011

From this chart we can see clear trends. First, both wage inequality has increased substantially from the 1980s to the present. The main reason for the rise in wage inequality has been the fall in wages for the lowest 20% of workers, not the rise in wages for other groups. The trend has also occured in Western Europe though the lower tier is becoming unemployed due to laws that prohibit wages from falling as far as they have in the US.

Possible Explanations for Rising Wage Inequality

  1. Decline in Unionization: many employees without a college education benefited from union membership into the 1970s. The decline in union influence means that many workers receive less compensation.

  2. Increase in International Trade: An increase in international trade most likely results in the US exporting technology-based goods and services while importing simple technology products. This means that workers in low-technology sectors face severe competition from abroad. Two implications from this are: 1. US workers who compete with imports see wages fall while wages in other countries rise, and 2. The prices of import-competing products fall in the US. Evidence on the falling prices of products is weak.
  3. Shift of Employment to Technology-Base: This results in a reduction in demand for less educated workers and an increase in demand for college graduates. Skills needed in the workplace are changing to require more analytical tools and more "brain-power."
  4. Increase in immigration: This impact mainly occurs because the number of relatively less-skilled immigrants to the US has increased, skewing the wage distribution downward.

There is wide disagreement among economists as to what is driving the wage inequality. To the extent that there is consensus, the shift in technology is believed to be the most important followed by international trade.