Constitutional Mischief: What's Wrong With Term Limitations by Nelson W. Polsby On November 6, 1990, California voters by a modest margin amended their state constitution with an initiative that promises to produce far-reaching consequences. This initiative, numbered 140, one of 26 statewide measures offered on the ballot, contained four main provisions:
(2) A lifetime ban on service in the state Senate after two terms, or eight years, and the state Assembly after three terms, or six years.
(3) A ban on the provision of a pension or retirement benefit for service in the legislature, other than federal Social Security.
(4) A sharp permanent reduction in the expenditures allowed for legislative salaries and staff, amounting roughly to an immediate 40 percent reduction in legislative staff.
- (1) A lifetime ban on the holding of each major state constitutional office applied to persons who have served two terms, covering the offices of governor, lieutenant governor, attorney general, secretary of state, treasurer, and superintendent of public instruction.
I shall not deal further with the first of these provisions, because limitations on the terms of executives having control of large and experienced professional staffs poses somewhat different problems for government -- though not necessarily for democratic theory -- than the set of provisions applying to the legislature.
While term limitations in some respects weaken all public officials in their incentives and capacities to attend to and effectively perform their duties, ballot measure 140 did special damage to the state legislature. Term limitation proposals meant to apply to legislatures elsewhere, such as the U.S. Congress, are designed likewise especially to cripple legislatures.
It can, of course, be argued that legislators have no need of pensions, staff, or longevity to perform their functions adequately. Did not Cincinnatus, after all, leave his plow only temporarily? Would we not be better off if the makers of our laws arose directly from the people, represented them faithfully, and then returned directly to the people before they can be sullied by the poisons lurking in the superheated political atmosphere of the capital?
There are at least two difficulties with this Grandma Moses depiction of modern American political life. The first has to do with the size and complexity of the districts being represented by individual legislators. The second has to do with the size and complexity of the legislators' tasks as authors of legislation and monitors of the executive branch of government.
One may doubt whether there is a legislative district anywhere that is small enough, or homogeneous enough, to be capable of producing a series of representatives to a legislature so perfectly attuned to the constituency's needs as to require no effort on the part of the member to ascertain what those needs are. Whether or not such a constituency exists in principle, we can be reasonably certain that large constituencies, containing half a million or more people, as congressional districts do, or 300,000-plus people, as California state Assembly districts do, require effort by representatives before these representatives can be expected to be adequate to the tasks of representation.
The belief in the spontaneous competence of good representatives belongs to an age and place where a relatively stable and well-accepted status system did much of the work of government. Amateur representation is historically the product of a status-ridden society, where people knew their place. People who inhabit more fluid and diverse societies, where people contend, negotiate, sue, migrate, or otherwise engage in mutually adaptive activities, require far more in the way of explicit attention from their representatives than Cincinnatus and his amateur colleagues are competent to give.
Translated into modern terms, the plea for an amateur legislature is a plea for legislators capable of representing only the most conspicuous elements of their constituencies. Perhaps this is all many modern legislators choose to do. But term limitations make legislators much less capable of choosing otherwise.
Equally serious are the consequences for the legislature as a body. Suppose we accept the notion, central to the term limiter's case, that state government, or national government, is somehow itself the problem. Would it not follow that it is in the interests of the people to equip their chosen representatives with the capacity to deal with the problem in all its complexity and difficulty?
The essential tool of a legislative body in dealing with any set of complex problems -- including those posed by the rest of government -- is its own capacity to form judgments about the merits of alternative proposals and diagnoses independently from whatever is advocated by the executive branch, the governor, lobbyists, fixers, hangers-on, and interest groups. The legislature's capacity to form an independent judgment arises from three things: a division of labor, in which individual members are assigned to specialize in a limited range of topics; longevity, in which they continue in their specialization long enough to familiarize themselves with the players, their interests, and the strengths and weaknesses of various arguments and proposals; and finally, their ability to purchase technical assistance responsive to their needs for analysis and understanding, in short, staff.
It follows that a constitutional amendment prescribing limits on the longevity of members of a legislature and limits on the amount of staff they may have to help them is an attempt to prevent the legislature from forming its own well-considered and independent judgment on matters its members must understand.
By no means, however, can we say that this exhausts the mischief enacted by California measure 140. We can say with some confidence that the capacity of the individual legislator to represent well has been deprofessionalized, and that the capacity of the legislature to act independently and competently has been severely maimed. But there is more.
Consider the social composition of a legislature that has been transformed into a body of temporary place-holders. To whom might membership be attractive? At least three sets of people: (1) the old (2) the rich (3) the bought.
A lifetime ban after six years of service and no pension can scarcely be a deterrent to a candidate who already has a pension, not to mention a short life expectancy. So we could expect that hale and hearty senior citizens might well enjoy the stimulus and the opportunity to serve that the people of California have devised for them. To be sure, there are obvious possible disadvantages to rule by the old. Elderly members of a legislature may be unable to work very hard. They may prove to be unusually stubborn, inflexible, or resistant to new ideas or new experiences. They may be out of touch with the vast bulk of the population they are supposed to be serving. Their health, or their hearing, or their physical robustness, may deteriorate on the job, making them unusually dependent rather than independent as policy-makers.
These are merely possibilities. Older legislators may also be wise, disinterested, experienced, and public-spirited. With respect to most of the human qualities we would want to see in a legislator, the old might constitute a population only somewhat more risky than the population at large. Of course, they could not be expected to be entirely free of concern for the special interests of the old.
It seems unreasonable to expect utter selflessness from any group in the population. Even so, I suppose more than a few might advocate generous provision for the school-age population, even if it meant means-testing benefits to the elderly. Considerations of this kind would fortify arguments favoring a mix of demographic characteristics in a legislative body that seeks to represent a diverse population rather than an absolute bar to membership by elderly members.
But consider the next category, the rich, to whom pensions and occupational careers, legislative or otherwise, would mean little or nothing. Why would they seek to serve at all, except to further their own interests? Some, no doubt, do serve for reasons other than purely to defend then-economic advantages. In Colorado, for example, where legislators are grossly underpaid, an unusually large number of well-to-do women, who do not need the money, can afford to serve, and many are reported to do so with distinction.
Nevertheless, it is very hard to justify a constitutional arrangement that systematically favors legislative service for a group so well able to take care of itself, if need be, in the private sector. And yet there can be no serious doubt that this is more or less the constitutional arrangement that Californians have voted for. The main sponsors of this plan in California were conservative Republicans and land developers, in short, friends of the rich. The main advocates of term limitations for members of Congress are Republican political operatives at the national level and trickle-down conservatives.
Finally, there are the overly dependent whom I have unkindly called the bought. We must recognize that term limitations greatly increase the dependence of those legislators who do wish to have careers, or have no extraordinary independent sources of income, on the money that interest groups can provide. Interest groups, many of which are currently allied with legislators, gain power over their legislative allies under term limitation conditions at three points.
First, interest group money is necessary to advertise new candidates for legislative seats in order to get them elected in the first place. Veteran legislators can depend on name recognition, habit, or constituent case work to reduce their dependency on contributions of cash. Indeed, well-entrenched incumbents can raise money more easily precisely because they can do more for interest groups than interest groups can do for them. Not so for novices.
Second, inexperienced legislators after they are elected need interest groups in order to help them understand political issues that come before the legislature. This is especially the case in the absence of legislative staff. Lobbyists have information. They are available to impart it. They offer alliances. They are indispensable to any legislative body with serious work to do.
But legislators must also have the capacity to keep interest groups at arm's length, so that those legislators who have the desire to carve out an independent path may do so. A reasonably safe seat, expectations of an honorable legislative career, a decent salary and pension, adequate staff, all play a part in providing that capacity. Without any or all of these things, interest groups gain influence.
What of the future? A legislator who knows that he has not much of a personal future in the legislature must consider in what arena his future lies. Will the member try for some other public office? Will the member become a lobbyist, or a lawyer in private or semi-public practice? The prospect of no future in the legislature does not mean no future at all. The California constitution does not now require that members completing their terms be taken out and shot. So here again the potentiality for undue dependence on interest group alliances raises its head. And this may do great harm to the quality of the public service that legislators are meanwhile able to render.
It may nevertheless be too severe to characterize legislators under a term limitation scheme as "bought" because the dependence of so many of them on interest group alliances and interest group money will inevitably increase. There will always be at least a few people of high character in most human institutions. In designing institutions, however, as we do when we write or rewrite constitutions, it is only prudent to organize incentives so that it is unnecessary continually to rely on individual resources of character and self-restraint to prompt good behavior. Rather, we should want our representative institutions to be positively hospitable to a great range of our fellow citizens, especially those leading normal lives, with everyday concerns, families to feed, ambitions and thoughts for the future, in short, those with as full a complement of human attributes and qualities and potentialities as we can attract to the messy and inspiring business of self-government.
Term limitations short-circuit this process by making it too hard for too many sorts of people to participate. This debases the representative capacities of the legislature as well as the capacity of the legislature to play a constructive independent role in governance.
White House Chief of Staff John Sununu, presumably speaking for the President, has recently endorsed the idea of a constitutional amendment limiting the terms of members of Congress. The absence of something like the self-destructive California initiative process at the national level means that it probably won't happen. James Madison and his colleagues knew very well what they were doing when they made the U.S. Constitution difficult to amend.
Observers meanwhile should reflect upon the spectacle of self-proclaimed conservatives trumpeting their support for a constitutional amendment to rid themselves of Democratic members of Congress whom they don't seem to be able to defeat the old-fashioned way. If alleged conservatives will launch a fundamental attack on basic American political institutions on so trivial a pretext as their annoyance at having to make bipartisan compromises with a coequal branch of government, what are we to expect of non-conservatives?
Those who remember Robert A. Taft s encouragement of Joseph McCarthy or George Bush's more recent attempt to rewrite the First Amendment to prevent flag burning may not be surprised when conservatives exhibit an appalling shallow-ness of commitment to basic American institutions. It may be argued that constitutional arrangements are so well settled in America that they can withstand any amount of self-serving, opportunistic rubbishing. This, however, is not the contribution American conservatives ought to be making to political discourse.