The Evolution of Social Security's Taxable Maximum

Summary

As of 2011, payroll taxes for Social Security are applied to the first $106,800 of an individual's earnings.¹ This taxable maximum (or "tax max") increases annually, according to growth in the national average wage index.² However, Social Security's projected funding shortfall has led some policymakers to propose increasing the tax max beyond the indexed levels to help restore financial balance. This brief does not take any position for or against modifying the tax max; instead it provides context for the Social Security reform debate by summarizing the changes that have occurred in the tax max and covered earnings since 1937.

Major Findings

• The tax max has been in place since Social Security's founding, but Congress has modified it over time to address several policy goals, such as improving system financing and maintaining meaningful benefits for middle and higher earners.
• Although the nominal value of the tax max has grown from $3,000 in 1937 to $106,800 today, in inflation-adjusted dollars the tax max declined from 1937 until the late 1960s, and then grew once it was indexed to wage growth in 1975. In wage-adjusted dollars, the tax max has remained roughly constant since the mid-1980s.
• The percentage of workers with earnings above the tax max ("above-max earners") fell from 15 percent in 1975 to about 6 percent in 1983 and has remained at that level since.
• Historically, an average of roughly 83 percent of covered earnings have been subject to the payroll tax. In 1983, this figure reached 90 percent, but it has declined since then. As of 2010, about 86 percent of covered earnings fall under the tax max.
• The percentage of earnings covered by the tax max has fallen since the early 1980s because earnings among above-max earners have grown faster than earnings among the rest of the working population.

Increasing the Tax Max: Historical Methods and Rationales

Although Social Security has included a tax max since its inception, it was not included in the plan drafted by President Roosevelt's Committee on Economic Security. That plan focused on poverty alleviation and would have exempted from the program nonmanual workers with monthly earnings of $250 or more. Instead, the House Ways and Means Committee instituted the tax max, setting the initial limit at $3,000 per year (per employer), equivalent to 12 months of earnings at the $250 level (Mulvey 2010). Mulvey finds that the reason for adding the tax max is not entirely clear from the written record of the Committee's proceedings. However, the administration's original exemption would have excluded high-income individuals (reducing income to the system which could be redistributed to low- and middle-income workers), limited the workforce covered by the program, and created what Mulvey terms "erratic" coverage for workers whose earnings fluctuated around the monthly threshold. Replacing the exemption with the tax max addressed these concerns, while still meeting the goal as a social insurance program of focusing on low- and middle-income workers who were more likely to be economically vulnerable in retirement.³

The tax max remained at $3,000 until 1951, when Congress increased it to $3,600 as part of the Social Security Act Amendments of 1950, which also expanded benefits and coverage in a variety of ways. Although the taxation elements of the legislation addressed the need to finance increased benefits, Social Security Commissioner Arthur Altmeyer argued for the tax max increase in the context of maintaining the relationship between benefits and preretirement earnings for middle and higher earners. Testifying before the House Ways and Means Committee, he suggested that "if the wage base is not raised, the differential between benefits for low-wage and high-wage workers will not adequately represent their differences in levels of living and the benefit structure will tend more toward a flat level" (Altmeyer 1949, 9). Congress increased the tax max to $4,200 in 1955, $4,800 in 1959, and $6,600 in 1965 to provide more meaningful benefits for middle- and high-income workers while also raising additional program revenue.

In 1967, the Johnson Administration asked Congress to substantially increase benefits and requested three increases in the tax max, to $7,800 in 1968, $9,000 in 1971, and $10,800 in 1974 (DeWitt, Béland, and Berkowitz 2007, 256). As before, these increases were intended to improve future benefit adequacy, meaning that benefits would more closely resemble preretirement income for higher-wage workers, while also financing higher benefits for current retirees. Congress did not grant all of these increases, but the 1967 Social Security Amendments increased the contribution and benefit base to $7,800 for 1968 (SSA 2010a). Although the other increases were not granted then, similar increases would be requested and granted during the Nixon Administration. Social Security Commissioner Robert Ball emphasized the increase in future benefits resulting from the change. He suggested that "only by increasing the earnings base can the program be kept up to date and continue to perform adequately for the average worker…a $3,000 increase in the earnings base in a 2 ½ year period has not been hailed as the major accomplishment in the program, but I think it's important to realize that in many respects it is the most important accomplishment" (DeWitt, Béland, and Berkowitz 2007, 265).

Although the tax max increase between 1965 and 1968 nearly doubled that of the previous three decades, a year later, President Nixon proposed increasing the tax max again and indexing it to earnings growth thereafter. Like Altmeyer and Ball, Nixon framed the tax max change as an issue of adequacy, saying "the goal of Social Security is the replacement, in part, of lost earnings; if the base on which contributions and benefits are figured does not rise with earnings increases, then the benefits deteriorate" (DeWitt, Béland, and Berkowitz 2007, 270). The 1972 Social Security Amendments increased the tax max to $9,000 in 1972, $10,800 in 1973, and $12,000 in 1974, with subsequent increases indexed to changes in the national average wage index (DeWitt, Béland, and Berkowitz 2007, 277; Ball 1973).^{4, 5} These amendments also substantially expanded benefits for current retirees, echoing past legislation in which tax max increases played a crucial financing role while simultaneously raising future benefits for affected workers.

Unrelated to changes in the tax max, the 1972 amendments introduced a "flaw" in the benefit formula that placed Social Security in an untenable financial position as benefits rose beyond their intended levels (DeWitt, Béland, and Berkowitz 2007, 318). The program also faced a general financing problem requiring policy adjustments. The 1977 amendments addressed the financing problem in part by wage-indexing initial benefits and by increasing the tax max between 1979 and 1981 by greater amounts than indexing alone would have yielded (Snee and Ross 1978). That is, ad hoc increases in the tax max exceeded those required by the automatic provisions of the 1972 law.

In his signing statement on the 1977 amendments, President Carter emphasized that beyond their importance in terms of increasing program revenue, the tax max changes were also significant because they achieved this goal in a manner that fostered equity. He suggested that addressing the shortfall primarily through increasing the tax max was "making the system more progressive and minimizing the added burden for low- and moderate-income workers" (Woolley and Peters, 2011).

Since the 1977 amendments were enacted, the contribution and benefit base has risen automatically with increases in average wages, largely addressing the historical goal of maintaining the relationship between preretirement earnings and benefit levels as wages rise.⁶ Current proposals to further increase the cap have instead tended to emphasize the rationales that framed the tax max portion of the 1977 amendments: reducing Social Security's projected funding shortfall, while creating a less regressive payroll tax structure, particularly in response to changing earnings distributions. Table 1 lists the tax max values from 1937 to 2011 along with a brief summary of the policy rationales for the changes and the mechanisms by which the levels were determined.

Value of the Tax Max Relative to the Larger Economy

Policymakers have used three common measures of tax max earnings values: nominal dollars, inflation-adjusted dollars, and wage-adjusted dollars. Chart 1 shows 1937–2009 tax max values for these three measures. The tax max values described above and shown in Table 1—in which the tax max was virtually static in the early years of the program and has grown steadily since the 1970s—are expressed in nominal dollars. The tax max did not increase between 2009 and 2011, as there was no cost-of-living adjustment (COLA) for recipients.⁷

Chart 1.
Nominal, inflation-adjusted, and wage-indexed tax max values, 1937–2009

SOURCE: Authors' calculations based on SSA (2010a, 2010b, 2010c) and BLS (no date).

However, the nominal values do not account for inflation. Inflating the nominal tax max values between 1937 and 2009⁸ to 2009 dollars using the Consumer Price Index for Urban Wage Earners and Clerical Workers (CPI-W) shows that the real value of the tax max fell in the early years of the program, before climbing intermittently during the late 1950s and 1960s to again attain approximately the real value seen in the program's first several years. Indexing the tax max to wage growth from 1975 onward has caused its real value to grow since then, as wages have generally increased more quickly than inflation.

The third approach is to adjust the tax max values for wage growth. This measure is especially significant because the tax max applies to wages, and has been indexed to wage growth since 1975. The wage-adjusted value uses the average wage index (AWI) calculated by Social Security's Office of the Chief Actuary. Wage-adjusted tax max values fell in the earlier years of the program before rising in the 1970s and early 1980s. Since the mid-1980s, the wage-adjusted value of the tax max has remained nearly constant at around $100,000, a predictable result reflecting the effect of wage indexing.

Proportions of Workers with Earnings Above the Max

The proportion of above-max earners in a given year has fluctuated since Social Security began in 1937. There were sizable changes in the percentage of covered workers with above-max earnings in the early years of the program, followed by a long period of relative stability (Chart 2). The proportion started below 5 percent in the late 1930s, rose to a high of 36 percent in 1965, then fell back to about 6 percent by 1983, where it has remained since, reflecting the consistent tax max indexing procedure used since the 1977 amendments were fully phased in.

Chart 2.
Percentage of covered workers whose earnings exceed the tax max, 1937–2009

SOURCE: Authors' calculations based on SSA (2011).

Chart 2 shows the annual percentage of workers with earnings over the tax max; however, the percentage of workers who earn over the tax max at least once during their lifetime is higher, as some workers exceed the max in some years and not in others. Using microsimulation data from the Social Security Administration's Modeling Income in the Near Term (MINT) model, we project that from roughly the 1951–1955 birth cohort to the 2011–2015 birth cohort, between 20 percent and 25 percent of individuals will earn above the tax max at some point during their working careers (Chart 3).

Chart 3.
Projected percentage of individuals who will earn more than the tax max in at least one year, by birth cohort

SOURCE: Authors' calculations based on MINT.

Historical Taxable Proportion of Covered Wages

Current policy discussions often focus on targeting the tax max to cover a specific share of economy-wide earnings, such as 90 percent of covered wages (Ball 2006).⁹ The 90-percent level is a common target because it was the ratio of taxable wages to covered wages attained in 1983 as a result of the 1977 changes. Since that time, the ratio has fallen. This trend may seem counterintuitive given that, as discussed above, the percentage of covered workers with earnings over this level has remained roughly constant. The reason the share of covered wages subject to the tax max has declined is that wages above the tax max generally have grown more quickly than wages overall.¹⁰

Although some suggest targeting the tax max to 90 percent of covered wages, others have argued that if historical precedent matters, looking at the program only after 1983 is insufficient (for example, Biggs 2009b and Tanner 2005). The average percentage of covered earnings subject to the tax max throughout Social Security's history is 83 percent, lower than the most recent annual estimate (86 percent in 2009). Chart 4 shows the ratios for 1937–2009, with the historical average highlighted.

Chart 4.
Percentage of covered earnings subject to the tax max

SOURCE: Authors' calculations based on SSA (2011).

Earnings for Above-Max Earners Have Grown Faster than Overall Earnings

To delve further into the trend in earnings distributions discussed above, we analyze the relationship between above-max earnings and below-max earnings using a tax-max ratio, defined as the percentage of earnings above the tax max divided by the percentage of workers who earn above the tax max. This ratio illustrates whether the earnings over the taxable maximum among above-max earners are higher (value greater than 1) or lower (value less than 1) than the average of overall wages for that year. The degree to which the ratio is above or below 1 illustrates the level of earnings inequality between above-max and below-max earners.

Chart 5 shows the tax-max ratio for 1937–2009. The lowest ratio was around 0.7 in the early 1950s, when the tax max was relatively low in relation to average earnings, and more than one-third of workers earned more than the tax max. Thus, relative to other periods, the earnings of above-max workers were not so different from those of workers who earned less than the tax max. By 1983, when 90 percent of earnings were under the tax max and about 6 percent of workers earned more than the tax max, the ratio had risen to 1.6. By then, the tax max was indexed to wage growth and only very high earners exceeded the tax max.

Chart 5.
Tax max ratio, 1937–2009

SOURCE: Authors' calculations based on SSA (2011).

The ratio has steadily risen since 1983, even as the proportion of above-max workers has stayed constant at about 6 percent. In 2000, the ratio was 2.72 (Chart 5), showing that earnings growth among above-max earners has differed substantially from that of the rest of the working population. However, this does not mean earnings have grown at the same rate for all above-max earners. The most substantial growth has occurred in the top 1 percent, and in particular, the top 0.1 percent of earners.¹¹

Conclusion

Social Security's tax max has evolved throughout the program's history. It started at $3,000 in 1937 and stayed at that level until 1951. Congress then raised the threshold, first through ad hoc changes and then through wage indexing, to accomplish various policy goals such as increased financing for contemporaneous benefit expansions and improved future benefit adequacy for middle and higher earners.

Since the passage of the 1977 Social Security Amendments, the tax max has increased steadily with the average wage. However, because Social Security faces a projected funding shortfall, many policymakers have discussed increasing the tax max beyond its wage-indexed values. Proponents cite the higher growth rate in above-max earnings relative to earnings beneath the threshold as a rationale for using the tax max as a mechanism to improve system financing.

The tax max's theoretical and political development, as well as its relationship with wages, inflation, and earnings inequality all play a role in the discussion of tax max modification. This brief provides background and context to help inform the policy debate.