Accountability and the Elected Official:
The Case for Pay-for-Performance for Congress and the President
Doug Koplow / February 1996
As hundreds of thousands of federal workers were furloughed without pay in December and January, the House of Representatives repeatedly blocked an amendment that would have cut their own pay until the impasse had been solved and a budget passed.1 Blocking this amendment is a microcosm of much that is wrong with our government. Cooperation for the common weal is nearly defunct. Fiscal constraint, in all of its glorious monotony, is sadly missing. While Congress and the President talk day-in and day-out about the need for deficit reduction, much of the most eggregious wasteful spending continues unabated. The Cato Institute's recent tally of how corporate welfare is faring in the appropriations process is telling. Despite the rhetoric from the ardent "small government" folk, support for our needy multinationals has been reduced by a scant 16 percent.2 And as the deficit grows, our elected officials feel practically no repercussions for the folly that they themselves orchestrate.
Our political system is premised on geographic representation and geographic accountability. While this structure is critical to our functioning as a Democracy, it is, by itself, insufficient to protect our common fiscal resources. There is little incentive for any individual member to give up benefits for the District to benefit the country as a whole. This lack of incentive costs us all dearly. Our tax revenues have become a commons; the profligate spending a tragedy. The time has come to end this tragedy of the commons by paying our elected officials based on how well they manage it for us.
The pool of resources that Congress and the President manage together is enormous. With outlays of nearly $1.5 trillion in 1994,3 federal spending was equivalent to the combined revenues of the 32 largest U.S. corporations that year.4 Federal outlays per U.S. citizen more than doubled in real terms between 1960 and 1990. At a whopping $2.7 billion in federal outlays per Congressperson per year, Newt's share of 1994 spending exceeds the revenues of Dow Corning; Dick Armey those of Polaroid. While much of these federal outlays are not discretionary, Congress and the President can (and do) exert substantial control even over the entitlement portions.
Since our elected officials control such vast resources, we should be more concerned with their inability to use our resources wisely. Deficit spending or not, the deficits should at least be buying us something useful. Yet, the lubricant of compromise in our current system is not common welfare or even common sense, but pork-barrel political trades. And, so long as the pork keeps flowing to Des Moines or Duluth, Joe Voter keeps supporting his Rep at the polls. Although Joe Voter talks a good game about being concerned about the amorphous "debt," he is not all that keen to give up his local perks to bring it imperceptibly lower, and Newt and Dick and Bill all know this. Tax-payer financed favor is traded for tax-payer financed favor to solidify the needed votes. The deficit grows, and confidence in the ability of Congress to govern the nation dwindles.
Let us begin to instill in Congress and the President a desire -- nay, a need as great as their drive to be on television at campaign time -- to spend public resources wisely. We can do this by linking their compensation to how well they do their job. A pay-for-performance system could help both to improve the accountability of elected officials and to restore some of the public trust in the institutions of Congress and the Presidency.
In determining how to establish such a link, corporate compensation systems offer useful guides. Although corporate compensation systems are far from perfect, the good ones do a fairly good job at making individuals with disparate interests work for the gain of a single organization. The analogy between corporate and public managers is a strong one: once the glory of introducing bills and presiding over endless committee meetings has been stripped away, the central function of Congress is that of a manager.
In a corporation, upper management determines the allocation of capital, the development of new markets, and the retreat from declining ones. Management makes high-level hires and high-level fires. When the company does badly, managers themselves may pay through reduced compensation (often in the form of declining value of the stock they own), or, in the extreme, with their jobs.
Members of Congress set tax policy and decide the direction of spending. "Market expansion" is done through new bills and taxes; "market retreat" through privatization, tax cuts, spending rescissions, or "Contracts with America." Congress approves high-level appointments, and exercises much power in their removal as well. In only one area do the tasks of public and private managers markedly differ: public managers are responsible for ensuring an acceptable level of public welfare as well as financial solvency.
In terms of accountability, each Congressional Member does face a job review every few years in the form of elections. However, unlike a firm, the body is rarely judged on its performance as a single entity, and no one, except for the President, is judged by the nation as a whole. Imagine a corporation in which every division competed to outspend the others, with little ability of shareholders to reign them in and no managers focused on company-wide performance. With Congress, so long as the part (the District or State) continues to perform well, most voters take little heed that the nation itself is slowly bleeding.
Any corporate manager responsible for effectively allocating nearly $3 billion in wealth per year would be compensated in large part based on how well those funds were deployed. Though the goals of a public manager differ somewhat from those of a private manager, the link between pay and performance can and should also exist in the public sphere. Linking pay to performance could improve the types of spending decisions that Congress makes, as well as the compromises made between Congress and the President. Individual members would think twice about serving up low-value pork if it meant their own salary would fall noticeably. Since all members would suffer lower pay if Congress as a whole spent resources foolishly, Members would begin to monitor the fiscal impact of each other's policy initiatives much more closely.
If fiscal matters were the only component of a pay-for-performance system, the human aspect of governance would be overlooked. Therefore, it is critical that indicators of social welfare be incorporated into the performance measures used to determine salaries as well. These social indicators could also force Congress to take more direct responsibility for making sure that government programs to alleviate poverty and hardship actually work -- whether they are in the form of bureaucracies or block grants. Pay-for-performance would not only give the two branches, and individual members, an incentive to work together to cut ineffective spending, but it would provide a quick and visible indicator to the country of how well these public officials were doing their job. Even the elected officials who are motivated by public service rather than financial rewards would pay heed to the signals such a system would send.
Implementing an incentive pay system involves three central issues: base pay; the criteria used to evaluate "good" versus "bad" performance; and how these criteria can be adjusted to ensure that Congress and the President strike a prudent balance between current spending and long-term well-being.
Low Base Pay With Large Incentive Pay for Jobs Well-Done
Rather than facing the awkward task of voting themselves pay raises every couple of years, each member of Congress, as well as the President, should have a starting wage equal to the average earnings of the American worker. In 1993, the average U.S. worker earned about $32,500,5 substantially below the $133,600 and $200,000 earned by the Congress and the President respectively.6 While lower than the base pay of most executives, this starting wage provides important common-ground with the electorate, and will ensure ample drive on the part of officials to earn their incentive pay. Certainly, this incentive pay should be generous: if they do a good job running the country, our elected officials deserve to be paid at a level comparable to that earned by executives at successful U.S. corporations the size of Polaroid or Dow Corning -- on the order of many hundreds of thousands of dollars per year. Yet, when they run the nation poorly, the average wage is more than sufficient a reward.
In addition, where institutional gridlock forces the government to shut down, fairness dictates that the management should suffer along with the staff, and Congressional and Presidential pay should immediately stop, with no retroactive renumeration. One must imagine that the recent stalemate would have been resolved far more quickly had Congress faced an unfunded Christmas along with the furloughed federal employees.
"Good Management" Applied to Congress and the President
The most critical aspect of making a pay-for-performance system work is properly defining what constitutes "good" performance. While fine-tuning such a system requires careful study, the core components of good performance are fairly easy to identify. Clearly, a "good job" running a nation is more complex than the bottom-line profit or return on investment measures used in the private sector. Therefore, performance measures for Congress and the President would have to blend financial indicators with social indicators. This seems possible.
On the financial side, current pay would be based on current spending, with larger deficits leading to lower compensation. Some may argue that deficits are not inherently bad, and that Congress should not, therefore, suffer reduced wages for running them. In fact, deficits represent a trade-off between current and future consumption, in the same way that corporate debt offers both risk of default for a corporation and potential rewards from expanded investment. Congress now has little incentive to strike the proper balance, or to ensure that deficits are used to benefit the country in the longer-run. Where deficits do serve a vital national interest, such as financing a war or pulling the country from recession, the wage reduction this spending would automatically trigger for the elected officials certainly seems equitable given the hardships facing the country at large. If deficits are needed to finance boondoggles in the districts of powerful members, lower resulting earnings for all in Congress would perhaps incite the timid to band together, challenge the great powers, and stop the waste.
On the social front, Congressional and Presidential pay would fall as unemployment and poverty levels rise. While the federal government can't prevent all economic hardship, federal policies and programs can have significant impacts on many individuals and on the economy as a whole. The linkage between pay and economic hardship among the populace would be based not only on national averages, but on differentials across states as well. If the nation, on average, was doing well while many in Alabama or Arkansas suffered in unemployment, pay for the President and Congress overall would drop. Only by including such regional disparities in the pay-for-performance equation can one ensure that the Members from larger, more prosperous states have an incentive to care about their smaller or less prosperous brethren.
By focusing on core social measures such as unemployment and poverty in determining compensation, Congress would be forced to target limited government funds on the programs it thought would be most effective in alleviating these ills. Policies that hurt the economic climate of the country would increase unemployment, leading to lower Congressional compensation. Federal fixes for these problems would also have to be better-targeted. For example, support for "job training" as a concept would not be enough unless the programs actually worked. Private training programs, where the provider earned no fee unless they successfully placed and kept trainees in jobs, might become more common. With jobs programs and many other areas as well, the focus of public support would have to shift from symbolic actions to effective actions -- from reports to results -- in order for Congressional and Presidential pay to rise above that of the average worker.
Balancing Current Spending Against Long-term Investment
Congress has often been accused of short-sightedness. Why worry if a program put in place this year leads to large defaults and taxpayer bailouts in ten years so long as the program helps you get re-elected next year? Private industry has long grappled with similar problems. Managers may have incentives to pump up current earnings to maximize bonuses, even if these actions represent poor long-term decisions. Similarly, investment decisions made today continue to affect the firm for many years or decades. The firm must somehow give current managers, many of whom will not be around for the entire period of impact, an incentive to make decisions that are good for the firm. This is done primarily through compensating managers with stock options or actual stock. Since the stock reflects the expected future earnings of the firm (discounted to current dollars), managers who make decisions that hurt the firm's long-term prospects will see the value of their own individual shares decline as well.
While there is no exact equivalent to stock options in the public sector, there are some possible proxies. "Withholding" and "pension linkage" are two. Under a system of withholding, a substantial portion of the incentive pay earned by Congress and the President would be paid out over five to seven years rather than immediately. During the ensuing five to seven years, better data on the cost to the nation of policies implemented by any particular Congress would become available. This would alleviate the risk of mis-estimating (or gaming) any of the many measurements of Congressional activity that would impact the size of outlays, revenues, or the deficit.
"Pension linkage" is an attempt to mimic private sector stock holdings by linking the value of Member pensions to the functioning of the programs they oversee. While such linkage may not be possible for all government programs, it seems fairly straightforward at least with government-run trust funds or insurance programs, where bailouts are clearly visible. The goal of pension linkage, as with private sector stock options, is to ensure that Congressional managers pay careful attention to the long-term health of the programs they are responsible for.
For example, a Congressperson on the oversight committee for Social Security would see his entire Congressional pension put into the Social Security trust fund. Were the trust fund to become insolvent, the Congressional representative would lose his pension. Were the fund to require a bailout, the Member would lose a portion of his pension. This newly invested manager would also have an incentive to ensure that any positive fund balances earned a reasonable return from the Treasury. If the Treasury yields were too low, such a manager might work to enable portions of the surplus to be invested in other low-risk investments that would yield a higher return.
Federal deposit insurance, such as through the Federal Deposit Insurance Corporation (FDIC) and the now-defunct Federal Savings and Loan Insurance Corporation (FSLIC), provides another useful example of pension linkage. During the 1980s and early 1990s, the taxpayer paid $200 billion (on a present value basis) to bail out failed banks.7 A host of factors contributed to this loss, not least of which were cuts in the number of thrift supervisors and a weakening of required capital standards. This allowed weak banks to remain in business, greatly increasing the cost of the bailout.8
Imagine if Members of the banking committees at the time had had their entire pension holdings put into accounts pegged to the health of the deposit insurance programs they managed. The greater the required bail-out, the more pension savings these Members would have lost. While the exact ratio of bailouts to reduced pension value requires additional specification, it is clear that if any reasonable linkage had been in place during the banking fiascos of this time, many Congresspeople would have lost their entire retirement savings. Fearing such a loss, key Members would have not intervened quite so readily to spare their banker-friends the indignity of government oversight or higher capital standards. They may even have advocated deposit-insurance premiums more in line with the lender's risk level -- either through the FDIC or by using private insurance markets.
Because current decisions can create the conditions that lead to large trust fund or insurance losses years in the future, it is important to ensure that public managers take a long-term view of their oversight responsibilities. Allowing Members to "cash out" of their linked pensions immediately on departure from either Congress or a particular committee would provide little incentive for them to make good decisions in the period immediately prior to their departure. Therefore, pension linkage would be liquidated slowly, such as over a five year period, reinforcing the connection between sound long-term decisions and the personal wealth of the Member.
There are no doubt other mechanisms, beyond those identified here, that could be used to link the pay of elected officials to their performance on the job. These types of mechanisms should be vigorously explored, with the most promising approaches implemented in short order. Unfortunately, debates over how much an elected official deserves to be paid and what laws can be passed to force the deficit down are far more common. Both are sorry substitutes, for they miss the central point that elected officials have little or no incentive to protect the common fiscal resources of the nation, though they have wide latitude in how to spend them.
In this critical respect, our system of governance is horribly broken. A day's work in Washington can change the social and business environment in the country for decades, affecting millions of people. As voters, these people can effect some change on who represents their state or district. Yet as citizens and taxpayers, we have very little leverage to make our elected representatives work together to spend the vast resources they take from us every paycheck wisely. Paying members of Congress and the President based on how well they manage the affairs of the nation would be a significant improvement over the current system, forcing increased cooperation at a national level and increased personal financial penalties to Members for doling out lavish pork or providing poor oversight of the programs entrusted to them.
These requirements could be over-ridden, but only with 60 votes of the Senate (rather than a simple majority). PAYGO expired in 2002. Attempts are renewal are trying to exempt tax cuts from evaluation, even though the Bush tax cuts of 2001 were a major source of renewed deficits. (CBPP et al., 4/20/04).
- 1"No Budget, No Pay" legislation was introduced by Senator Barbara Boxer of California. It was passed by a voice vote in the Senate three times (9/22/95, 10/27/95, and 11/28/95). Identical legislation was introduced in the House by Representative Richard Durbin of Illinois. Each time, the House stripped the language from broader legislation, effectively blocking its passage. ("No Budget, No Pay" timeline; and Press Release, "House Republicans Sink `No Budget, No Pay' -- Again," January 31, 1995. Provided by the Office of Senator Boxer, February 1, 1995).
- 2Dean Stansel, "How Selected Corporate Welfare Programs Have Fared in the FY1996 Appropriations Bills," Cato Institute, December 11, 1995.
- 3"Federal Budget Outlays - Defense, Human, and Physical Resources, and Net Interest Payments: 1940 to 1995," Statistical Abstract of the United States 1995, Table 520.
- 4"The Fortune 500 Largest U.S. Corporations," Fortune Magazine, May 15, 1995, pp. F1 - F22.
- 5Internal Revenue Service, "Individual Income Tax Returns: Selected Income and Tax Items for Specified Years, 1975-1993," Statistics of Income Bulletin, Fall 1995, p. 142.
- 6Salary information taken from The World Almanac and Book of Facts, 1994 and 1996, (Mahwah, NJ: Funk & Wagnalls), 1993 and 1995, respectively.
- 7Bert Ely, "Government Regulation: The Real Crisis in Financial Services," in David Boaz and Edward Crane, editors, Market Liberalism: A Paradigm for the 21st Century, (Washington, DC: Cato Institute), 1993, p. 119.
- 8R. Dan Brumbaugh Jr. and Robert E. Litan, "The S&L Crisis: How to Get Out and Stay Out," The Brookings Review, Spring 1989, p. 7.