Introduction-

My central thesis is that there is a significant piece missing from virtually all current policymaking models. The missing piece is the business of someone in the policy process endeavoring to serve as an “honest broker” in the policy process. That is, there is in every complex policy issue a factual substrate that one can get right or wrong and that can be depicted accurately or in a misleading way. Disputes over this factual core–and attempts to distort and “spin” this core in self-interested ways–are a central dynamic in policymaking. This central dynamic is missing in virtually all policy models, or it is submerged in the model inside some other theoretical construct.

One of the procedures that effective policymaking must involve is in some way distinguishing between facts and opinions. This involves such matters as clarifying the scope of a problem; providing a balanced accounting of the pros and cons of various policy alternatives; explaining policy ideas and options objectively, and the like. Too often, these essential parts of the policy process are undertaken by the very parties-at-interest who are contesting over the policies. This leads, almost inevitably, to a confusion between facts and opinions in the policymaking process. Normatively speaking, the biggest need in the policymaking process is often for someone to play the role of the “honest broker” and to assist in getting the facts straight.

To account for this overlooked aspect of the policymaking process, I have developed an alternative policy model, which I call the Honest Broker (HB) policy model. I construct my model on the general scaffolding of a standard “policy stages” analysis. I argue that in each of the stages of the policymaking process, as identified by Kingdon, Anderson, and others, there is an opportunity for someone to play the role of the honest broker; that this role is sometimes played but is overlooked; and that (normatively) this role ought always to be played. I will argue that the reason this part of the policymaking process does not appear explicitly in the major policy typologies is because, while the factual core is always at issue, there is not necessarily always someone striving to fulfill the role of the “honest broker.” This role is too often absent from the policy process, which is why I think it is so overlooked in existing models, and is why there is a need for a normative aspect of this model.

I present a simple case-study of this model in action through the example of the current policy debates over the reform of Social Security. In this case study, my application of the model will be limited to the first stage of the policymaking process (Problem Definition). However, I will argue for the general applicability of my model to all stages of the policymaking process.

 

The Fact/Value Distinction-

Most of us understand the importance of the fact/value distinction in the analysis of public policy, and in broader domains of intellectual life as well. We have been sufficiently schooled in the realization that many alleged statements of “mere fact” are permeated with value judgments. Thus, one cannot wholly resolve most complex public policy disputes by appeal to “the facts” of the matter since ideologies, belief systems, biases, and other value-laden factors always figure in any complex conceptual question.

However, the observation that most factual claims contain value statements, does not imply that they have no factual content. That is to say, it is not the case that everything is a matter of value preferences just because most propositions contain value judgments. Some things are matters of fact.

Perhaps the classic exposition of this distinction for policymaking is to be found in the work of Herbert A. Simon. As Simon has observed: “. . . to assert that there is an ethical element involved in every decision is not to assert that decisions involve only ethical elements . . .  most ethical propositions have admixed with them factual elements.”[1] Even if every conceptually-rich proposition is a mixture of facts and values, we can, with some effort, partially disentangle these two elements.[2]

So for purposes of the HB model, it is not necessary to deny or dispute the fact/value distinction. It is only required that we refuse to embrace a certain vulgar misconstrual of this distinction: the extreme idea that everything is a value judgment and nothing counts as a fact.[3]

 

The Problem with Current Policy Models-

Too often we model policymaking by constructing our typologies from the point of view of the already committed, that is, from the perspective of the various contending parties-at-interest in a particular domain of policy.

So, for example, Cobb and Elder identify only four sources of issue creation: 1) Groups competing over the allocation of resources; 2) Persons or groups manufacturing a issue in hopes of personal gain; 3) “Circumstantial reactors” who adopt an issue because of the press of events (e.g., an oil spill); and 4) “Do-gooders” who are motivated by some altruistic ideals to take up an issue.[4]

Rochefort and Cobb, while describing in some detail the five parts of what they call the “problem description” process, represent all aspects of this “anatomy” as being determined almost entirely by various forms of social construction, without any real acknowledgment of a factual core to any of the questions they pose.[5] For example, they say that the “Nature of the Problem” has five parts: Severity; Incidence; Novelty; Proximity; and a Crisis or non-Crisis quality. Yet none of these five attributes seem to them to have a grounding in fact. For example, when discussing Severity, they can manage only to observe that it is often a contested domain, and they use the misleading example of global warming to prove their point.[6]

Deborah Stone, as another example, designs her policy model in terms of the “causal stories” that are constructed around policy issues, and she depicts the policy process as being one of struggle between competing interests to shape the type of causal story being told so as to maximize their self-interests.[7] Schneider and Ingram see public policies as having targets, and the only questions of interest concern the degree of power possessed by those target populations and whether the social construction of their identities is positive or negative.[8] Even Theodore Lowi’s classic model of the policy process depicts policies in terms of their coercive force and the manner in which they distribute or redistribute the advantages controlled by government, thereby implying that grubbing for self-interest is the underlying dynamic driving the policy process.[9]

Finally, we can observe that even the broadest models of the policymaking process—such as the “stages model” used Anderson—make no explicit place for any kind of objective fact-finding as part of the policy process.[10] At best, this activity is submerged somewhere in an unacknowledged manner in some or all of the policy stages. For example, Anthony Downs has described a five-part “issue-attention cycle” in which interest in issues rises and falls in a fairly predictable way, but without this dynamic being obviously sensitive to the facts of the case.[11] And while Kingdon goes to great lengths to identify the various parties in the policy process (both inside and outside of government) one will not find in his descriptions any actor performing the role of the Honest Broker.[12]

What is typically overlooked in all these models is the simple possibility that there may be portions of the polity who have no interest at stake, or who have not yet taken a position on the issue, nor are they seeking the satisfaction of some altruistic impulse, but who are merely seeking to become informed about the issue so as to decide what in their judgment is the best public policy. Moreover, there may be actors in the policy process whose efforts are aimed at nothing more than helping the undecided members of the polity become better informed about the policy options before them. Virtually all the models of the policy process that we have are designed as if all the players are motivated only be various forms of self-interest, and as if the question of truth or falsehood in policy advocacy never seems to arise. So these two roles—that of the uncommitted member of the polity and that of the honest broker—are nowhere explicitly accounted for in our existing policy models. And yet, I would suggest that in most policy issues, the uncommitted far outnumber the self-interested. And while there may not be as many people striving to play the role of the honest broker as there are striving to manipulate the polity, this role too does exist and is missing from our policy models.

For example, those of us who are not “environmentalists,” or devotees of business for business’ sake, are puzzled on what is the best public policy regarding drilling in the Alaskan National Wildlife Reserve (ANWR). We would like nothing so much as someone to explain honestly and objectively what the pros and cons are of opening ANWR to oil exploration so that we might know what policy position we want to support. But an uncommitted citizen will seek in vain for such an honest and objective explanation.

Perhaps nothing is so frustrating to someone in such a circumstance (being open but uncommitted) as the encounter with the cacophony of self-interested voices clamoring to persuade a beleaguered decisionmaker of the merits of their own positions. All too often, not only is the undecided citizen presented with conflicting value claims and political ideologies, but even basic factual indicators of the size, scope and valence of the problem are in dispute. Committed advocates for interests-at-issue will too often “spin” even the factual substrate of a policy issue in ways designed to persuade decisionmakers to support their position. This is done through a variety of devices: selective reporting of factual data; false reporting of factual data; unstated assumptions; Hobbesian choices; a deliberate effort to muddle value judgments and facts together in a way that is designed to mislead, and so on.

So what we typically do as beleaguered decisionmakers is to collect the views of multiple advocates from various places along the ideological spectrum and try to somehow overlay them like multiple film transparencies one on top of the next in the vain hope that from their overlapping interference patterns some coherent policy picture will emerge. So we read the Washington Post and the Washington Times. We go to the Whitehouse website and then to the website of the Center for Budget and Policy Priorities. We read the views of the Century Foundation and the Cato Institute. We try to decide who to believe, since they so often make contradictory claims. This decision usually involves identifying whose side we are on ideologically or politically, and then accepting their representations of the policy issue as informing our own understanding. This cannot be, it seems to me, an ideal way to make sound public policy in the context of a democratic society.

I suggest that this tendency to “spin” and to mislead by obscuring the facts around policy issues is one of the most salient features of contemporary public policy debates. The necessity for a step in the policy process that counters this “spin” is probably one of the most pressing needs in any effort to make better public policies. We might call this effort to clarify the facts of a policy issue in an objective and honest way—and to “un-spin” the issues—the effort to serve as an Honest Broker in the policy debate.

If all this is so, then this dynamic—of “spin” and “un-spin”—must be accounted for in some explicit way in our policy models. That it is either absent or submerged in other theoretical constructs in most existing models is a key shortcoming of those models.

 

A More Complete Policy Model-

Sound policymaking requires clarity of understanding, among other traits. The tendency of the various policy players to “spin” the policy debate obscures the issues and merges and muddles facts, opinions, value judgments, and motives together into a kind of conceptual mélange that makes rational decisionmaking harder than it has to be. Policymaking would be better served by inscribing our policy models with an explicit step or stage in which someone playing the role of the Honest Broker endeavors to untangle the conceptual muddle and to lay out for decisionmakers the pros and cons of various policy options, and the assumptions underlying the analysis. This will not lead to perfectly rational decisionmaking, there is no perfection of that sort this side of heaven. But it will lead to better decisionmaking than the form we suffer under at present.

Making the Honest Broker step explicit in our policy models will thus serve both a descriptive and a proscriptive function. Descriptively, it will force policy theorists to search for actors playing this function and to describe their role more explicitly. Proscriptively, when no one is playing the Honest Broker role in a given policy dispute, the absence of this helpful step will be noted, thereby increasing the likelihood that someone will step forward and assume this responsibility. At a minimum, when the absence of an Honest Broker is noted, this can serve the quite useful function of putting us on our guard a bit more when considering the policy products of the “spinmeisters.”

The Honest Broker role ought to be active in all five stages of the policy process, as described by Anderson, either as a regular step in the process or as a separate policy stage itself.[13] There are two functions for the Honest Broker: identifying the facts in the policy issue and clarifying the values and assumptions involved. This is an interactive process potentially applicable in each stage of the policy process. This is depicted graphically in Figure 1.

 

Figure 1: The Honest Broker Model of the Policy Process

 

The use of the HB model occurs through the application of three specific procedures:

Step 1: Clarify what is fact-based and what is value-based in any given policy topic.

This step is of course not necessarily easy, and often it will itself be a “contested domain.” But even if the teasing-out of facts from opinions and values is not perfect or complete, some effort along these lines has to be better than nothing, and certainly better than the efforts of the advocates who deliberately conflate, mislead and “spin” these distinctions. Perfection is not required for the HB model to work, either in this step or the subsequent ones. “Better than nothing” will, I think, be a whole lot better than what we have now.

Step 2: For the factual components of the policy issue, identify instances of mistaken or misleading use of facts, and clarify the objective facts to the maximum extent possible.

The assumption here is that if the polity chooses policy options based on a false understanding of the facts in the matter, this is unlikely to lead to optimum policies. I cannot prove this thesis at this general conceptual level, but it seems a plausible presumption.[14]

Step 3: For the value-based components of the policy issue, point out the value commitments, and clearly identify their implicit and explicit assumptions and conceptual consequences.

The idea here is to clarify the often unstated and unexamined assumptions and implications of certain policy positions. The purpose of this analysis is not advocacy. It is not to challenge or change anyone’s value choices, but only to clarify them so that adherents can be sure they actually wish to embrace those value-positions. The premise here is that all too frequently we choose policies based on misrepresentations of the values in dispute, or a misunderstanding of those values.

The principle thesis of this 3-step procedure is that if a policy issue is systematically subjected to this analysis, that this will result in more-informed policy choices, and hence, better policy outcomes. The intuitive idea here is simple enough: ignorance and confusion are not a sound basis for making public policy. The aim of this procedure is to lessen both the ignorance and the confusion so that better policy choices can be made.

 

Where Are the Honest Brokers?

We might expect that in debates over problem definition, or in other stages of the policy process, that the media would play the role of the honest broker. Sometimes the press does play such a role, and sometimes it does not. For example, the press has recently done a fairly good job of clarifying the factual background to the present debate over the filibustering of judicial nominations. In this debate, the Republicans have depicted the Democrats’ attempts to filibuster judicial nominations as being without precedent in Senate practice, and hence the so-called “nuclear option” of banning filibusters on judicial nominations is seen as an appropriate remedy. In-depth examination of this issue by the press has revealed that the Republicans’ representations hinge on narrow technical differences in the tactics that they have deployed in blocking roughly the same percentage of President Clinton’s judicial nominations versus the tactics the Democrats have used in blocking President Bush’s nominations.[15] The claim that the Democrats’ behavior is unprecedented is thus factually false, and to the extent that members of the public may have been inclined to support the Republican policy agenda on grounds of fair-play, some may wish to alter their policy preferences.

Whether or not the press plays the role of the honest broker, the role still needs to be played. That the press sometimes fills this role but often does not is, in part, I suggest, because of the absence from our policy models of an explicit stage in the policy process where the role of honest broker is defined and expected.

The type of policy scrutiny and explanation that I am suggesting here is not entirely missing from the current policy domain. Various policy entrepreneurs have sensed this same need and launched efforts at the kind of analysis I am pointing to as essential for sound policymaking.

For example, the online Internet service FactCheck.org was created by the Annenberg Public Policy Center at the University of Pennsylvania precisely to serve this policy function.[16] Another example is the Pew Research Center, which describes itself as “fact tank” rather than a “think tank,” and which takes no position on any policy issues but strives, rather, to provide “information in the public interest.”[17]

The whole rise of organizations representing themselves as “nonpartisan” and seeking to provide balanced factual policy analysis is a phenomenon that is quickly becoming central to the policymaking process, and yet our policy models as yet make no explicit provision for this aspect of the policy process.[18]

So the concepts I am offering here are not unheard of or strange. They ought to be, I suggest again, a central feature of all good policymaking, and they must, therefore, be a central feature in our policy models. Let us consider, then, one simple example of the application of this aspect of policymaking to the current debates on Social Security reform.

 

A Brief Case Study: Social Security Reform-

During the 2000 presidential campaign the Republican candidate, George W. Bush, signaled his intention to introduce some form of Personal Savings Accounts (PSAs) as part of a reform of the existing Social Security system. In May 2001, as President, Bush chartered a bi-partisan commission[19] whose task was to suggest ways to introduce PSAs into Social Security. At the beginning of his second term, the President announced that his version of Social Security reform would be one of the top priorities of his administration, identifying it as one of his top four policy agenda items.[20]

In January 2005, the President launched a “60 cities in 60 days” barnstorming tour of the nation, which had two objectives: 1) to make the case that the Social Security system faced a massive crisis (this was the effort to make the case for action); and 2) to convince the public that his preferred solution centered on PSAs was the best policy response to the crisis. Although the effort to sell the country on the PSA policy fix has stalled somewhat, the President and his advisers have claimed credit for having convinced the nation that there is a “crisis” in Social Security financing.

Central to any definition of a policy problem, as Kingdon emphasized, is to identify and define key measuring indicators.[21] At the first of his town-hall meetings, the President defined the problem for his audience in these terms: “You realize that this system of ours is going to be short the difference between obligations and money coming in, by about $11 trillion, unless we act. And that's an issue. That's trillion with a "T." That's a lot of money, even for this town.”[22] Consistently, the Administration has insisted on using the $11 trillion figure to define the problem. This $11 trillion figure is also the number that members of the Republican Conferences in the House and Senate are instructed to use whenever discussing Social Security.[23] Democratic politicians and opponents of the President’s PSA plan consistently put the value of the program’s financing shortfall at $3.7 trillion, or prefer to use other measures entirely, such as the shortfall expressed as a percentage of taxable payrolls (see discussion below).[24]

Both of these figures might well be taken as objective indicators of a serious financing problem facing the Social Security system. But the larger the value that is accepted as defining the problem, the more willing the public will be to accept more “extreme” policy options in response. The creation of PSAs as part of the Social Security program is a policy proposal without precedent in the 70-year history of Social Security.[25] Thus defining the problem as of unprecedented scope helps rationalize this dramatic policy proposal. The $11 trillion figure is especially impressive as a problem definition since this figure is roughly the size of the nation’s annual Gross Domestic Product.

Although the President’s 2001 Commission issued a thick Final Report, neither this report nor any of the various background materials made available by the White House, have ever attempted to clarify the $11 trillion figure, how it is derived, what assumptions it is based on, and what alternative measures are available, and what the pros and cons of the various measures might be.[26] Engaging the HB policy process, we can clarify matters to a considerable degree.

The Social Security system has two sources of income: current payroll taxation and assets invested in two Trust Funds. To plan the management of this massive social insurance program, each year a Board of Trustees issues an annual report, prepared by the actuaries at the Social Security Administration, which estimates the financial transactions of the system.[27] Since 1965, these estimates have utilized a 75-year estimation period.[28] The program is said to be solvent if its projected income meets or exceeds its obligations, on average, over this 75-year period.

To make this assessment the actuaries must predict the future in great detail for each year of the estimation period, including such factors as: birth rates, death rates, immigration rates (legal and illegal), price inflation, productivity rates, future unemployment and wage levels, known policy changes, and literally dozens of other key variables. Attempting such long-range projections is an enormously uncertain enterprise, and although the estimates by the Social Security actuaries are the state-of-the-art in the profession, they can be off by orders of magnitude, even for very short periods, when there are significant policy changes. For example, the estimate issued in 1955 projected that total program expenditures would be $123 billion in the year 2000, whereas actual expenditures in 2000 were $408 billion—an error of 330% over that 45 year period.[29]

In the 2003 Trustees Report the Board of Trustees directed the actuaries to begin including a second long-range estimate in addition to the traditional 75-year estimates. The actuaries were directed to add an “infinite horizon” set of estimates which projects program financing for 995 years into the future. [30] Thus, the 2004 Trustees Report includes a set of estimates that project the various factors involved in Social Security financing for each year from 2005 up through the year 3000.

There are several traditional measures of financing adequacy, the most widely-used being how much tax rates would have to be raised or lowered in order to balance income and outcome. This measure expresses a deficit or a surplus as a percentage of taxable payroll, which is to say, as the adjustment that would be needed in the tax rate to achieve balance. In the 2004 Trustees Report, for example, the program was in deficit by 1.89% of payroll.[31] Simply put: to restore the program to actuarial balance, the payroll tax would have to be raised by 1.89% in 2005 to pay full promised benefits during the next 75 years.

Another measure of program financial health uses a concept called present value. The present value is a way of quantifying in present dollar terms the value of a fund over a long period of time—it is a way of taking into consideration the time-value of money. The use of a present value metric is itself a policy choice in that it tends to shift the focus away from tax rates and on to a large dollar figure, which can then be described as the program’s “unfunded liability.”

Expressed as a present value, the 75-year shortfall in Social Security’s financing is $3.7 trillion. Expressed as a present value, the 995-year shortfall in Social Security’s financing is $10.4 trillion, which the President and the Republicans in Congress round-up to $11 trillion.[32] Hence, there are two different figures for the value of Social Security’s unfunded liability, depending upon the period of time covered by the estimates.

This is the factual content of this particular conflict in problem definition. What, then, are the values and assumptions being used in the two sets of estimates?

I think it fair to say that the value-judgments involved here is that those who want to make the Social Security financing problem look as small as possible, will choose to use the $3.7 trillion figure, whereas those anxious to make the problem seem as big as possible will use the $11 trillion figure.

As to the assumptions being implicitly accepted: First, one must assume that the present value metric is the appropriate measure of the program’s financial health. Then, in order to accept the $3.7 trillion figure one must assume that the traditional 75-year period is the appropriate measuring period. To accept the $11 trillion figure one must assume that the 75-year estimates are too short, and one must accept the idea that the Social Security Administration’s actuaries can meaningfully predict the future course of the key indicators each year for the next 995 years into the future.

Thus we find that hidden in these two alternative claims about the size of the Social Security financing problem are the following sets of values/assumptions:[33]

 

 

I do not suggest that this process somehow magically yields an “answer” as to the correct policy choice. And, no doubt, there will be policy actors who will each insist on choosing their preferred measure regardless of any clarifications an honest broker might offer. But there are almost certainly also many non-committed decisionmakers for whom a simple clarification such as performed here might help them in deciding which policy agenda to support. The only role of the Honest Broker is to clarify the choices decisionmakers face. One may very well wish to embrace a policy agenda that is furthered by the use of an actuarial estimate extending 995 years into the future, but certainly one should not do so unknowingly.

Although a simple clarification such as this cannot determine a policy choice, I will conjecture that this type of clarification adds real value to the policy process because I am fairly confident that there will be a class of decisionmakers who, upon learning of the implicit values and assumptions behind the $11 trillion figure, will refuse to except this as the proper measure of the problem and who may therefore be less inclined to support the President’s preferred policy solution. If this is so, then the role of the Honest Broker can in some instances have a real impact on the decisionmaking process.

Now, having clarified the facts and separated them out from the value judgments and assumptions, we have performed the tasks of the Honest Broker for one very simple piece of the policy debates over the future of Social Security. Of course, any significant policy choice involves many more factors of problem definition than this simple case study. This same analysis can, however, be extended throughout this policy debate to dozens of key factors in the policy complex. Doing so, helps clarify the policy choices that face decisionmakers, and this type of analysis could easily be extended to other examples of Social Security policymaking, or indeed to any domain of policy subject matter.[34] 

This brief case study applies the HB methodology to only the first stage of the policymaking process. However, there is nothing in the model that limits its applicability to only this stage. The model aspires to be generally applicable to all stages of the policy process as all stages involve some degree of confusion and uncertainty as between facts and opinions or values. Hence all stages of the policy process ought to benefit from an application of the HB model. However, it may well be that, say, problem definition/agenda setting might be a more fruitful venue for the role of the Honest Broker than, say, implementation. But this thought remains to be explored in future work.

As a general proposition, it may also be that the HB model is a more effective tool in policy areas that are rich in factual content and fertile with ideological differences. So, utility rate-setting as a policy domain might be less in need of an Honest Broker than policymaking in the area of social welfare. However, this speculation too is a thought that shall have to be held in abeyance pending further case study research efforts.

 

Conclusions-

The foundation principle of democratic theory is the assumption that a well-informed citizenry is the best possible basis for making good public policy. Thus we need to take seriously—as modelers of public policymaking—this fundamental dynamic of democratic decisionmaking. Informing the public, honestly and objectively, is a core requirement of a democratic society and should therefore have an explicit place in our policy typologies.

This business of getting the facts straight is essential to the making of sound policies. Yet almost all of the prevailing policy models either ignore this aspect of policymaking or treat it as submerged in some other theoretical variable. I suggest, on the contrary, that “getting the facts right” is a central part of the policy process. To present a complete picture of the policymaking process we need to make an explicit place in our models for the identification and resolution of the factual core that is part of all policy disputes. Moreover, we need to explicitly identify the role of the “honest broker,” point out when it is being played, and, normatively, urge that it ought always to be a feature of sound policymaking.

Futhermore, I would suggest that existing policy models are becoming increasingly outdated to the extent that they do not have an explicit placeholder for the honest broker function in the policy process. As policymaking has grown in complexity, and as sources of information have exploded, the general public has become inundated with a growing multitude of competing claims about virtually every aspect of any given policy area. The polity is drowning is a sea of competing and incompatible claims. Spin is everywhere and the non-ideologue is often at a loss as to who to believe.

In some ways, I think this is the central and most salient policy-related dynamic of our time, and the theorists of the policy process act as if nothing has changed, and the models they first constructed in the 1960s are still as relevant as ever. I think, on the contrary, that these traditional models miss what is arguably the most important development in the policymaking process in our present day. The greatest need the polity has in the current policy environment is for someone without any axe to grind to help clarify our policy choices and to separate the facts from the spin. The Honest Broker model is a small effort to begin improving our existing models to better describe the policymaking process as it exists in 21^st century America.

 

---

References

Anderson, James E. (2003). Public Policymaking. Boston: Houghton-Mifflin. Fifth Edition.

Annual Report of the Board of Trustees of the Old-Age and Survivors Insurance Trust Fund. (1955). Washington: U. S. Government Printing Office.

Annual Report of the Board of Trustees of the Old-Age and Survivors Insurance and Disability Insurance Trust Funds. (2004). Washington: U. S. Government Printing Office.

Bush, George W. (2005). President’s State of the Union Address. Retrieved May 7, 2005. from the White House Web site: http://www.whitehouse.gov/stateoftheunion/2005/

Cobb, Roger W. and Elder, Charles D. (1983). Participation in American Politics: The Dynamics of Agenda-Building. Baltimore, Maryland: Johns Hopkins University Press.

DeWitt, Larry (1999). The History and Development of the Social Security Retirement Earnings Test: Special Study #7 by the SSA Historian’s Office. Retrieved May 7, 2005 from the U. S. Social Security Administration Web site: http://www.socialsecurity.gov/history/ret2.html

DeWitt, Larry (2003). A Brief History of Social Security.  Retrieved May 7, 2005 from the U. S. Social Security Administration Web site: http://www.socialsecurity.gov/history/brief.html

DeWitt, Larry (2005a). It Was Not About Race: The Decision to Exclude Agricultural and Domestic Workers From the U. S. Social Security Program. Under submission to Social Science History.

DeWitt, Larry (2005b). An Examination of the Transition Costs of Personal Accounts in the Bush Plan for Social Security Reform. Paper written as part of course requirements for Public Policy 610E, Spring 2005 term, University of Maryland, Baltimore County.

Downs, Anthony (1972), Up and Down with Ecology: The Issue-Attention Cycle. Public Interest, XXXII (Summer 1972), 38-50.

Gibson, Gail (2005, May 15). Senate Split on Stalling Tactics. Baltimore Sun, p. A1.

Kingdon, John (2003). Agendas, Alternative, and Public Policies. New York: Longman. Second Edition.

Lowi, Theodore (1972). Four Systems of Policy, Politics and Choice. In McCool, Daniel C. Public Policy Theories, Models, and Concepts (pp. 181-201). Upper Saddle River, New Jersey: Prentice-Hall.

Rochefort, David and Roger Cobb, (1993) Problem Definition, Agenda Access, and Policy Choice. Policy Studies Journal, 21, 56-71.

Saving Social Security: A Guide to Social Security Reform (2005). House and Senate Republican Conferences, January 27, 2005. Unpublished internal party document, copy in author’s possession.

Schneider, Anne and Ingram, Helen (1993). How the Social Construction of Target Populations Contribute to Problems in Policy Design. Policy Currents, 3 (1).

Simon, Herbert A. (1997). Administrative Behavior: A Study of Decision-Making Processes in Administrative Organizations. New York: The Free Press. Fourth edition.

Social Security Forum (2005). President Participates in Conversation on Social Security Reform. Andrew W. Mellon Auditorium, Washington, D.C., January 11, 2005. Transcript of President’s remarks. Retrieved May 7, 2005 from the U. S. Social Security Administration Web site: http://mwww.ba.ssa.gov/history/gwbushstmts5.html#01112005

Stone, Deborah (1989). Causal Stories and Formation of Policy Agendas. Political Science Quarterly, 104 (2), 281-300.

Strengthening Social Security and Creating Personal Wealth for All Americans (2001). Report of the President’s Commission to Strengthen Social Security. Washington, D.C., December 2001.

Wade, Alice (2005). [Author’s interview with the Deputy Chief Actuary for Long-Range Estimates at the U.S. Social Security Administration, 4/3/2005.]

Ways & Means Democrats (2005). Challenges Facing Social Security. Document by Democratic members of the House Ways & Means Committee. Retrieved May 7, 2005 from House Ways & Means Committee Web site: http://www.house.gov/waysandmeans_democrats/issues/social_security_challenges.shtml