In Dye's analysis, little emphasis is placed on the role of individual wealth in providing access to power. If anything, Dye makes a sustained effort to play down the role of individual wealth. Yet, even in his own terms, the overwhelming majority of elites are recruited from the upper middle and upper classes. Furthermore, the foundation of the oligarchical policy making model that he provides is upper class wealth, that wealth controlled by the upper 2 and a half percent of the wealthy. The money to finance groups such as the CFR, the Business Roundtable, the Council on Economic Development and so forth is all upper class wealth, and as Steve Brouwer commented in Sharing the Pie, these are the "institutions of solidarity that the rich have created to protect themselves and their assets." (p. 48). Consequently, it is necessary to examine the class structure and the distribution of wealth and income, if we are to understand the pathways to power.

     To begin to think systematically about the distribution and the exercise of power let us first address the manner in which, and the extent to which, class relations reproduce themselves. That is, do the rich stay rich and the poor stay poor? Are the rich powerful, and how does wealth translate into political power?

     I begin, then, with a basic assumption: one of the most important, if not THE most important power resource is wealth. Later we will go into other resources, but wealth is important because it places an individual with wealth in positions of power visavi those the individual employs, those to whom the individual lends money, and those to whom the individual gives money, particularly politicians.

     Perhaps the most fundamental point to begin with is that what most of us take to be the political arena is actually only the tip of the iceberg. Elections and public office holders are generally not the movers and shakers in the political world, but rather, politicians are the foot-soldiers of the ruling class and elections are little more than the means of legitimation. Now obviously, that is a contentious statement and one that requires considerable elaboration, but for the moment let me simply ask you to suspend judgment on the statement and accept it as a basic operating assumption.

     Secondly, the ruling class and the upper class are by no means isomorphic. Indeed, the ruling class is generally a small subset of the upper class. Dye estimates that the ruling class is approximately seven to eight thousand people, Prewitt and Verba estimate four to five thousand leaders. It is interesting to note that Dye, who is generally associated with the power elite model, has a higher estimate than Verba and Prewitt who are generally more closely associated with the pluralist model.

     The first step in building a model of power, then, is to identify and estimate one of the most important sources of power: wealth.

     How do people acquire property and wealth?

     Life-cycle earnings and inheritance are the two major ways of acquiring wealth. The question, then, is, of all the wealth in the U.S., how much has been acquired through inheritance and how much through earnings during the life-cycle?

     Let us address this issue by building a model. {graph here}

     The specific predictions of the life-cycle model are that net worth will increase with age and that the distribution of net worth will be more equal within age cohorts than within the general population. Under this model, the inequality of property ownership which we observe is explained as due to a combination of individual tastes, lifetime earnings, and the natural cycle of life through which we all pass.

     As tests of the theory, one must ask:

     a) Can it explain the total capital stock and the volume of private savings which we observe?
b) Can it explain the inequality of wealth which we observe ?
C) Is it in accordance with the savings behavior which we actually observe among individuals?
d) Is it true that wealth inequality within cohorts is less than that in the population as a whole?

     White did a simulation study in 1978 in which he predicted that AT BEST about 42% of annual private saving in the U.S.. can be accounted for by the life cycle model. Kotlikoff and Summers estimated in 1981 that AT BEST 19% of total U.S. wealth can be explained by a life cycle model. Note that these are the BEST cases. In terms of with in cohort differences, Atkinson concluded in 1975 that the concentration of wealth within age groups is not markedly less than in the population as a whole. It is true that average wealth increases as individuals age, but the life-cycle model cannot explain the large fraction of the population who have very low net worth at all ages.

     But that is not to say that the life-cycle model has no applicability. Indeed, the life-cycle model does in fact explain the behavior of about 95%, or indeed 97.5%, of the population. The non-rich, that is 95% of the population do fit the model fairly well except they do not "run down" their wealth at the end of the life cycle. The model does not fit particularly well a number of sub-groupings within the 95% however. For example nonwhites, rural families and non-high-school graduates do not in any way match the predicted model. So the only group whose behavior is modeled at all well by the life-cycle model is urban, college-educated whites --- and then only if one looks at holdings of housing, durable goods and cash. Over the population as a whole there is simply not enough individual savings from earnings, on average, to generate the amount of capital which we observe. The evidence rules out life-cycle lump savings as the major determinant of capital accumulation in the U.S. economy. Since the life-cycle savings model fits neither the few who are rich nor the many who are poor, it is probably best to think of it as a model of middle-class behavior.

     Instead, it appears that the massive intergenerational transfer of wealth is the major source of capital in the U.S. economy.

     Essentially, their are two types of wealth which are passed from one generation to the next: real salable property, and "human wealth". "Human wealth" refers to the value of an individual's potential future earnings which one must work for in order to realize the potential. So, for example, part of human wealth passed from one generation to the next is in the form of education.

     The question then, is how much of an impact will inheritance produce on the wealth distribution of the next generation. Thus we must ask:
a) How much wealth is passed from generation to generation?
b) To whom are inheritances left?
c) How are inheritances combined in current families?
d) What is the rate of accumulation of inherited property?

     Most of the research from both radical and neoclassical perspectives conclude that:
a) Most people will receive from their parents either a zero inheritance of property on an inheritance composed of depreciated consumer durables and/or housing.
b) The children of more wealthy families will receive an inheritance partly in the form of human capital and partly in the form of property.

A critical factor is how inheritances are re-combined. Here we need a term: "Assortative mating" which refers to the tendency for individuals to marry people from similar social backgrounds. Perfect assortative mating implies that individuals always marry people with the same amount of family wealth. If children of well-off families marry beneath themselves, family fortunes are fragmented over generations and wealth will tend to spread more equally over time. In fact, the correlation of socioeconomic status scores between the fathers of U.S. couples was found to be between .39 and .7 in 1973.

     Among the top 100 or 200 wealthy families in the U.S., inheritance is the overwhelming source of wealth. At the very top, for example, in 1982, when FORBES published a list of America's billionaires, thirteen of the fourteen on the list had inherited all or most of their wealth, and eleven of the thirteen family fortunes derived from oil.

     Given that one's initial "stake" is perhaps the most important barrier anyone entering business must overcome, it is clear that inheritance provides an important leg up on the competition. Thus, while many individuals work hard and improve upon their inheritance, without their inheritance they would be salaried employees who would have to depend on their life-time earnings in order to accumulate enough capital to begin a business late in life. Thus, as Thurow has noted, families in the top 5% or 10% of the wealth distribution may move up much more easily if inheritance gives one generation the start it needs in business.

     Now, if an economy is a competitive one, opportunities for extraordinary profits as in oil are essentially transitory, but if barriers to entry prevent, for a time, outside competitors from joining the industry and thereby diluting the bonanza, much larger fortunes may be acquired.

     Thus an important variable is the degree of monopoly in any given economy. In 1975, Commoner and Smiley estimated, under different monopoly assumptions, that between 40% and 66% of the wealth of the top .27% households is capitalized monopoly profits and between 20% and 50% of the wealth share of the top 2. 4% is similarly due to the presence of monopoly.

     Lets take a closer look at the distribution of wealth. As Osberg notes in his excellent study of economic inequality in the U.S. (quoting Jones), "`For a general statement, it would seem that the most likely pattern of change in the United States as a whole over the past two centuries was 1) some increase in wealth inequality to 1860-70, 2) possibly continued to a somewhat higher degree of concentration, which may have peaked around 1890, or 1929, or 1940, 3) a mild downdrift in inequality to the 1950's and 4) little change" (p. 47) until 1980. In the past two decades, the trend toward increasing concentration of wealth has returned. By way of a snap shot in three different centuries, in 1774 the top 10% owned 45% of all wealth, in 1870 they owned 66% of wealth, and in 1983 they owned 69% of wealth. (Brouwer, p. 42)

     There are several summary measures used to assess the degree of income and wealth inequality. The most straight forward is simply to report wealth and income by quintiles, sometimes breaking the top quintile into 10%, 5%, 2 and a half %, 1% or .5% because wealth and income are particularly concentrated at the very top. But this method requires five or six numbers and is often difficult to compare across time. Perhaps the most common single measure is the Gini index, created by Corrado Gini in 1912. The Gini coefficient measures the degree to which the distribution of income deviates from perfect equality. (draw graph on p. 17 of Osberg)

     The drawback of the gini index is that societies that have similar average income and gini coefficients can have very different income distributions. The gini index can only be used as a comparative figure if the distribution curve of one country or point of observation lies wholly within the distribution curve of the other country of point of observation. Furthermore, the gini coefficient is more responsive to changes in distribution among the middle classes and is not as sensitive at the extremes.

     Fortunately, there are other, albeit less common, measures of inequality that behave in different ways. For example, the coefficient of variation focuses on the inequality between the top and middle of the income distribution, while the Atkinson measure emphasizes either the high or the low end of distribution but should always be used instead of the Gini index if the Lorenz curves of two countries or time frames intersect. A fourth measure of inequality is the Theil index, which focuses on group differences and if one is comparing groups, it is the appropriate index. (p. 27 Osberg shows all indexes and compared to several countries)

     Unfortunately the other indexes are not as frequently employed as the Gini index, so most studies report the Gini coefficient. Here are some examples. Between 1983 and 1989, the Gini coefficient for the distribution of wealth in the U.S. rose by 2.1 percent, indicating more inequality in the distribution of wealth. "Of more significance is the actual size of the Gini number for both income and wealth. In [1980 the Gini coefficient for money income in the U.S. was .4 (Osberg, p. 17)], in 1983, the Gini number for income distribution was .414; the number for wealth distribution was .777. By 1989 the Gini number for income distribution had risen to .428, and for wealth distribution to .793." Peterson, p. 111.) In 1969, Wolff estimated the gini index of personal disposable wealth to be .72, indicating a trend toward greater inequality of wealth between 1969 and the present Likewise, between 1970 and 1990 the Gini index for income in the U.S. rose by 12.1 percent, indicating a substantial increase in inequality. The greatest shifts, however occurred since 1980.

     Looking at percentages gives an alternative picture: "In 1977, the top 1 percent of households got 7.3 percent of the income, whereas in 1983 the top 1 percent owned 31.5 percent of the nation's financial wealth. By 1988, the top 1 percent had increased their income share to 12.8 percent, and their share of net worth to 37 percent." (Peterson, p. 112.) Just as significant is the nature of the wealth owned by the top ten percent compared to the bottom 90 percent. "According to information in the 1992 Green Book (published by Ways & Means), 66.5 percent of the wealth owned by families in the lower 90 percent of families was real estate, and of this the overwhelming proportion (85.5%) was equity in the family home. Money in checking and savings accounts, CDs, and money market funds accounted for the next-largest form of family wealth (13.3%), followed by automobiles (6.8%). Families in the lower 90 percent had only 3.4 percent of their financial assets in stocks and bonds...In 1989 the top 1 percent of families owned 49.4 percent of common stocks and 78.9 percent of bonds, public and private. The lower 90 percent owned only 14.6 percent of stocks and even less of bonds, 6.5 percent. Of business assets, 61.9% were owned by the top 1 percent and another 29.1 percent by the next 9% of families, leaving only 9% owned by the lower 90% of families." (Petersen, p. 113.) In short, for the overwhelming majority wealth is not income producing but in the form of fixed capital or inaccessible, primarily homes, cars and pension funds. For the wealthy, 61% of income is unearned investment income. (Brouwer, p. 7.) But that is probably an underestimate. In a 1976 study Sawyer found that "American statistics succeed in counting 98.3% of wage and salary income, 91.1% of entrepreneurial income, and 45% of property income in the income distribution statistics.(Osberg, p.25) In another study in 1981 by Radner found much higher rates of under reporting. He found "property income (interest, dividends, rent, royalty, and trust) to be underestimated by a whopping 135%. Since property income is highly concentrated among the more affluent, he concluded that the Current Population Survey underestimated the share of the top 5% by about one-tenth and overestimated somewhat the shares of the fourth through ninth deciles." (Osberg, p.25) Physical assets compose the bulk of the bottom 90%, while financial assets compose the bulk of the top 10%'s wealth. (p. 41 Osberg gives table of wealth)

     If wealth provides the foundation for upper class entrance into positions of economic power, and if economic power is a major stepping stone to political power, then we must also study the distribution of income since for the overwhelming majority of the population the only economic resources available for use in the political process derive from personal income. This is not to say that there are not other types of resources available to the non-upperclass. For example, numbers of people and their ballots, to some degree people's time, and in some cases motivation, can counter at least partially the economic resources of the upper class. But the potential each of these resources represent, clearly, is limited by economic resources.

     For example, if communication of information is a strategic resource in political discourse, it takes money to produce newsletters, produce mass mailings, rent buildings for large meetings and conventions, pay for the organizations secretarial needs and so forth. Some of these needs can be met by donations and volunteers, but of course that is a much less secure base than the ability to secure these services through the ability to pay for them.

     So, obviously, the distribution of income places some limits on the potential for political mobilization and political participation. Again, I remind you that virtually every theorist writing about democracy has emphasized the close relationship between the equal distribution of resources and the possibility of democracy. The more equally resources are distributed, the greater the possibility for democracy. It is not a guarantee that democracy will manifest itself, but equality provides the requisite conditions for the growth of democracy.

     So, what is the distribution of income in the U.S. both at present, and over the past few decades? Since the 1930's there has been relatively little change in the distribution of income.

     The poorest 20% of the population has received between 3.1 and 5% of total income since the mid-1930's. But it was only in one year, 1945 that the poorest 20% received 5%. From 1975 to 1980, for example, the poorest 20% of the population received 3.8% of income in each year except 1975 when the figure was 3.9% according to the Bureau of the Census.

     The second quintile, 20%-39% of the population has received between 9.7 and 10 .6% of income in every year since 1935 except in 1947 when the figure was 11.0%.

     The third and fourth quintiles are closest to their expected income based on an equal distribution of income. The third quintile has. received between 14.1% and 17.6% of income in every year since 1935 The fourth quintile has received between 20.9% and 24.9% Of income in every year since 1935.

     Finally, the Top 20% of the population received 51. 7% of income in 1935, but in every year since it has ranged between 43.6% to 48.8% of income. In other words, the top 20% receives more than twice the proportion of income than would be expected if incomes were distributed relatively equally. If we look at the top 5% the ratio moves from 2 to 1, to 3 to 1. That is, the top 5% of the population has received between 16.6% and 18.7% of income in every year since 1947.

     Well, next week I will put this data into a larger context, but in the meantime I would like you to think about the significance of this data in terms of the exercise of power in the U.S., the possibilities for channels of participation in the political process for ordinary people, and the impact these dynamics have on the possibility of democracy. Next week we will devote an hour to a general discussion of these issues.