Policymaking for Social Security is a wonderful book—perhaps the best book ever written on American social policy. Derthick's achievement is all the more remarkable when it is considered as the middle book of a “social policy trilogy,” along with Uncontrollable Spending for Social Services Grants (1975) and Agency Under Stress (1990). Taken together, these books provide as fine an introduction to American social policy as I can imagine; taken together with The Politics of Deregulation (1985), which Derthick wrote with Paul Quirk in this same period, they provide an incomparable overview of American politics.

Policymaking for Social Security conveys as few other books do the complexity and richness of American politics. It is not surprising that post-modernists, proponents of rational choice approaches, and devotees of path dependence all see elements of their preferred approaches in this book, for these forces are all present in her book, just as they are in real life. Indeed, what is most interesting to me about Policymaking for Social Security is Derthick's analysis of the complex mix of motivations of politicians and program executives, and how preferences and “zones of acceptable outcomes” can change over time in response to changing political conditions.

Thus, Policymaking for Social Security is a kind of political science Rorschach test, in which we bring our own analytical baggage to bear on an empirical reality that simultaneously incorporates and transcends our limited analytical perspectives.

I will focus here on what I see as a central puzzle raised by Policymaking for Social Security and, more specifically, on the reason that Derthick gives for writing the book. At the end of the first chapter, she argues that Social Security “policymaking has long been characterized by constricted participation, doctrinal rigidity, and extreme inertia” (12). She argues that “the prime need is to enlarge the possibilities of choice” (12) and states that her goal is to raise the extent of dialogue among an informed citizenry (13).

Of course, she refers to the opening of a dialogue as a hope rather than as a prediction. But a striking feature of the past 20 years of Social Security politics is that a substantial broadening of the dialogue about parameters of change along the lines that Derthick called for has occurred. In particular, calls for incorporating individual investment accounts into Social Security have been spurred by critiques from conservative policy intellectuals like Martin Feldstein and from think tanks like the Heritage Foundation and Cato Institute.

Along with the broadening of the dialogue, there has been a long-term change in the program's political underpinnings. In the developmental period that Derthick focuses on in Policymaking for Social Security, for example, top program executives mostly came up through the ranks of the Social Security Administration (SSA), and were supporters of incremental expansion of the program. These top program executives virtually monopolized agenda-setting for Social Security. In the contemporary period, the top leadership of SSA generally has come from outside the program, has been less vocal and visible, and has for the most part not played a leading role in the Social Security debate. It can no longer be said of program executives that they “define the terms of debate” (405). In Congress, policy specialists on the Ways and Means and Senate Finance Committees formerly exercised a virtual monopoly on Social Security policymaking; the closed nature of policymaking was exemplified by the closed rule granted Social Security legislation in the House. The political dynamic of congressional policymaking was driven by competing desires to claim credit for benefit expansions and to protect the program from unsustainable demands. In the current period, on the other hand, there is competition for control of Social Security policymaking among party leaders, expert committees, and Budget committees, with blame-avoiding and blamegenerating considerations paramount.

Great transformations are evident outside of government as well. Organized labor has been supplanted by seniors lobbies (notably AARP) as the main interest constellation involved in Social Security policymaking. External critics of Social Security in its development phase were based in the insurance industry and politically marginalized; their recommendations mostly posed the unattractive alternative of means-testing benefits. Currently, external critiques are concentrated among powerful voices on Wall Street and in think tanks and advocacy groups with strong links to the Republican party, offering the seemingly very attractive option of individual accounts with (until a few years ago) high stock market returns.

All of these changing conditions, except for the interest group environment, might be expected to have led to significant policy change over the past two decades, or at least to have brought more issues up for sustained congressional debate and definitive legislative resolution. But despite the broadening of the dialogue and changes in the constellation of political forces around the program, there has been virtually no change in Social Security's main features since the “rescue bill” of 1983, which stabilized the programs finances in the short- and medium-terms through a combination of benefit cuts, taxation of benefits for upper-income recipients, and a gradual increase in the age at which full Social Security retirement benefits are received from 65 to 67. Benefit taxation for highest-income recipients was increased in 1993, and the Social Security earnings test was abolished for workers aged 65 and over in 2000. But these are very modest changes. Fundamental reforms remain far from adoption in the U.S.—indeed, they have not even proceeded to the stage of serious congressional debate.

The virtual absence of policy change in Social Security policy is particularly striking if it is contrasted with the public pension systems in other advanced industrial countries over the past 20 years. In countries like Italy, France, Germany, Sweden, and the United Kingdom, a variety of retrenchment and refinancing measures—and in some cases more fundamental restructuring—have been recurring items on the policy agenda over the past two decades. The pension reform debate in Europe has also given much more attention to mechanisms for giving pension credits to care-givers (especially for young children) than in the United States.

Why have a broadening of the terms of Social Security dialogue and major changes in the Social Security policymaking environment not led to a broader and more sustained congressional debate and an unfreezing of policy? There are several reasons. One is simply demographic: the ratio of workers to retirees has not declined nearly as far in the U.S. as in most Western European countries, nor does as high a percentage of American workers retire in their late 50s and early 60s.

A second set of reasons concerns policy feedbacks from prior policy choices. In particular, the distinctive U.S. policy choice of a “Bismarckian Lite” pension regime—social insurance based, but with relatively low replacement rates—has turned out to be politically and financially sustainable for longer than most European pension systems. Current Social Security surpluses have kept incremental retrenchment initiatives off the agenda, while long-term insolvency projections and the requirement that Social Security be funded entirely by contributions have prevented a recurrence of the bidding wars of the early 1970s. The maturity of the Pay-As-You-Go system poses huge transition costs for any shift to a funded system of individual accounts, keeping it off Congress's “action agenda” if not the broader “discussion agenda.” The non-sustainability of budget surpluses have further weakened prospects for a system of individual accounts. In short, the Social Security system has evolved in ways that, in the medium term at least, have made reform less immediately necessary and have made critics' proposals less practical.

A third set of reasons concerns the interest group environment. AARP is by far the largest of the seniors' organizations, claiming a membership of more than 35 million Americans age 50 and above—nearly three times the membership of the AFL-CIO. AARP's membership includes a broad age and ideological spectrum: only half of its members are retired, and approximately one-third are under age 60. The large size and ideological diversity of its membership make it difficult for AARP to take a positive stance in favor of any particular direction for policy change. When it has done so—as with the Medicare “Catastrophic” benefit in 1988 financed by an income-related premium on the elderly, and more recently with a prescription drug benefit—the organization encountered significant resistance from its grassroots. But AARP is very wellplaced to resist policy initiatives that attempt to impose losses on all or part of its constituency. Thus the constellation of interest pressures, like other mediating forces in the United States, militate strongly against major changes in Social Security.

Given this policymaking environment, are there opportunities for the sort of informed and sensible dialogue that Derthick recommended in Policymaking for Social Security, and for alternatives to the current ideological and partisan impasse? One can certainly imagine the elements of a reform package that would bridge the current rift. It might include trading-off an increase in the payroll tax, with all of that tax devoted to a system of individual accounts, packaged with a gradual lowering of current guaranteed replacement rates, increased credits for care-giving, and an improved income floor for those with the lowest incomes. In the short run, however, such a compromise seems very unlikely. The two parties' elites and political bases are very polarized, and both parties see Social Security as an issue that may work to their electoral advantage. Thus the outlook for Social Security reform in the United States remains one in which blame-generating and stalemate will likely continue to dominate over the next decade.

The politics of Social Security are likely to shift again when Social Security expenditures begin to outpace contributions in a little over a decade. But the likelihood of huge public debts accumulating over the next decade will constrict restructuring options further, and the political weight of the elderly and near-elderly will only grow stronger. Barring a change in the current bitter and relatively even partisan divide among the mass electorate and political elites, the prognosis is for a debate that is both informed and acrimonious, and for reform that is both too little and too late.