Stanley Bach "Mandates, Consensus, Compromise, and the Senate*"

Papers on Parliament No. 48
January 2008

Prev | Contents | Next

The paper that I am discussing with you today grows out of interests that were reflected in my 2003 book on the Australian Senate, Platypus and Parliament.[1] More immediately, the themes I will develop here have been stimulated by the paper that I prepared in anticipation of my current visit to the ANU’s Parliamentary Studies Centre.[2]

That paper, which I shall revise with the benefit of comments I hope to receive from some of you here today, examines the fate of Senate amendments to government bills during the past eleven years. Using data gleaned from the Senate’s annual publication on the business of the Senate,[3] I have attempted to trace the path that each Senate amendment took until its ultimate disposition. My purpose has been two-fold: first, to develop some empirical evidence about how often the Senate, when it had a non-government majority, evidently compelled the government to accept amendments to its legislation that it would not accept voluntarily; and second, to investigate how this record of Senate legislative influence has or has not changed since 1 July 2005, when the government assumed numerical control of the Senate for the first time since 1981.

I will not review my findings in detail now because the full text of the paper will be available for any of you interested in reading it. With regard to my first purpose in looking at the documentary record, I will summarize a more complicated set of findings by saying only that I did indeed find evidence of the Senate’s influence on legislation, manifested in amendments that the House of Representatives initially rejected but that ultimately either were adopted or that led to other changes in government bills that both houses accepted. With regard to the second purpose, I will observe simply that, through the end of 2006 at least, the change in partisan control of the Senate has been reflected in the virtual disappearance of Senate amendments that the government opposed but that nonetheless led to changes in government legislation. Although I have not yet had the opportunity to incorporate data for this year, it already is clear that the 2004 Senate elections did make a difference in the government’s ability to work its will in this building.[4]

I doubt that these summary conclusions will come as a surprise to anyone here today. But social scientists like to think that there is some value served by documenting the obvious, because every once in a while, what we think to be obvious proves not to be. There remain, however, two related questions that the documentary record cannot answer. During the years of non-government control between 1996 and mid–2005, was the Senate’s impact on legislation a little or a lot, and was that impact good or bad? Answers to these questions depend not so much on our interpretation of any data, but on what we think the Senate should do—how it should exercise its powers and responsibilities as a participant in the national legislative process. This judgment, in turn, depends on how we believe national policy decisions should be made under Australia’s constitution or under any other reasonably democratic constitution. There are few questions more fundamental to the governments under which we live, whether in Australia or the United States.

Before going any further, let me acknowledge the bias that I bring with me today. In 2000, in connection with the Commonwealth’s centennial commemorations, Professor Elaine Thompson, who introduced us to the concept of Australia’s ‘Washminster mutation’ of the British and American models of government, wrote a paper for the Parliamentary Library on the first century of Australian parliamentary democracy. In her paper, she commented on how the how the role of the Senate has changed in recent decades. She concluded by writing that:

[t]here is still no convention concerning the limits of the Senate’s powers with respect to the Executive. Indeed it is reasonable to suggest that there is a political convention developing which expects the Senate to play a restrained, but nonetheless active role as a second chamber reviewing and, on occasion, rejecting government.[5]

This ‘restrained, but nonetheless active role’ is what I had in mind when I entitled the lecture I gave here in early 2003 ‘The Delicate Balance’ in the Australian political system. I think it fair to say that there is no dearth of people in this building—especially those who walk on blue carpet, many who walk on green carpet, and even some who walk on red carpet—who will willingly and enthusiastically make the case for restraint. It should come as no surprise, on the other hand, when I say that, as an American who spent most of his professional life working for the U.S. Congress, I prefer to make the case for activism.

Let me summarize the argument I am about to make by saying that there are at least three basic models of the decision-making process that can characterise the way in which democratic—or shall we say republican?—regimes work. One of those models is, to my mind, flawed on both empirical and normative grounds, especially when applied in the sweeping way that its proponents often advocate. Another of the three models is a chimera whose advocates are in need of a radical platitudectomy. As you already will have guessed, it is the third model that I intend to advocate as being best suited to preserving and improving the long-term health of my society and perhaps yours as well; but it also is a model that is under threat in my country and seemingly discredited in yours.

I will proceed by discussing each of these models in turn, asking whether any one of them is more likely than the others to characterise how democratic governments can and should work. In the process, I also will make some comments about the implications of my argument for the United States, and I even may venture some thoughts about what it might mean for Australia.

Mandate Majoritarianism

The first model that I want to discuss briefly posits that, by the very definition of what constitutes a representative democracy, government policies are to be decided by the majority of representatives who, in turn, speak for and act on behalf of a majority of the electorate. To the extent that constitutions prescribe decision-making by majority vote, this may seem to be little more than a truism. Yet majority control of policy inevitably is constrained—by constitutions, as they may be interpreted by courts; sometimes by requirements for super-majorities for some purposes, such as to override an executive veto of legislation, as in Washington; and always by the inescapable need for unelected officials to make policy themselves in the process of filling in the interstices of enacted law.

Furthermore, the kind of majoritarianism I have in mind goes beyond the recognition that democratic constitutions typically give a majority of MPs the power to work their will. The more ambitious form of majoritarianism, what I will call mandate majoritarianism, begins with recognising the power of the parliamentary majority, but then moves on to contending that the majority has the right and the obligation, as well as the power, to control parliamentary decisions because that is what the voters expect and demand. At the moment, my argument is that mandate majoritarianism rests on a collection of assumptions about the behaviour and motives of the electorate for which there is little empirical support. Later I will take the argument further by suggesting that this model of decision-making would be undesirable even if there were an adequate empirical basis for it.

I critiqued the notion of electoral mandates at some length in Platypus and Parliament, so I will content myself today with summarizing that critique and the empirical reasons for doubting the existence of any such mandates.

In summary, the argument for mandate majoritarianism is this: as each parliamentary election approaches, each of the parties contesting it puts forth a catalogue of specific and detailed programs that it will enact and implement if entrusted with the power of government. Each voter then studies and evaluates all of these catalogues (otherwise, platforms or manifestos), selects one because he or she endorses all its elements, and then votes for that party. One of the parties receives a majority of votes that translates into a majority of seats in the parliament, or in the house that monopolises power if it is a bicameral parliament. That party then enacts into law, and without substantive change, the programs enumerated in its electoral catalogue of promises, and once those programs are enacted, the governing party proceeds to implement them in the manner it deems most consistent with the commitments it had made to the electorate.

Where this argument goes beyond recognising the parliamentary majority’s power to control parliamentary decisions, if that majority is sufficiently unified and determined, is in contending that the majority has both a right and a responsibility—it has a mandate—to do so. In entrusting a party with a majority of seats in parliament, the electorate thereby gives that party not only the power but also the right to enact and implement its program because the election results constitute a blanket endorsement of that program. Moreover, this is more than a grant of discretionary authority to act. The governing party is obligated to enact and implement its program, again without substantive change, because a failure to do so would constitute a breach of trust with the electorate. The party must do what the people have elected it to do. It not only has been given a mandate to govern, it has been mandated to govern.

Two elements of this argument may seem unreasonable: first, that the party must enact and implement each and every one of the policies and programs it advocated during the election campaign; and second, that it must enact and implement them without substantive change. You may think that these are unnecessarily restrictive requirements. Upon closer examination, though, it should be clear that removing these elements from the argument would leave the majority party, once in power, with so much discretion that the individual voters could have no real confidence that they really knew what package of policies and programs they would be ‘buying’ if their party were victorious. If that party, once elected, could pick and choose from among the promises it made during the campaign, or if it reserved to itself the right to embody a general campaign promise—to ‘control the growth of government’, for instance—in any one of a myriad of legislative forms, the linkage between election and governance would be too weak and unpredictable to be very meaningful.

It should be noted that the potential applicability of mandate majoritarianism is limited. All else aside, it makes sense as a model only when the nation’s electoral system is very likely to produce single-party majorities in government or, as in the case of Australia, the possibility of a majority composed of a stable and durable coalition. This is most likely though not inevitable when MPs are elected from single-member constituencies. Such systems tend to encourage two-party competition—and, therefore, a clear electoral victor—because the election in each constituency is a winner-take-all contest in which parties that are unlikely to win have trouble convincing voters to support them. (Preferential voting weakens the strength of this argument, of course.) Those systems also tend to give the party winning the most votes more than its proportionate share of parliamentary seats, thereby increasing the likelihood that there will be a single-party government.

In presidential-congressional systems, there is the possibility of divided partisan control, with one party controlling the executive and another controlling one or both houses of the legislature. In fact, this has been a common condition in the United States during the past half-century. Under these circumstances, no party can make a convincing claim for a mandate to govern, although American presidents routinely do so when confronted by a Congress in which the other party has a majority in the Senate or the House of Representatives. So too in parliamentary systems in which a single party, constituting the government, does not hold a majority of seats in parliament (or in the house of parliament that matters). If there is a minority government, and non-government parties and independents comprise a majority in parliament, the governing party has no reasonable claim to an electoral mandate. By the same token, if there is a majority coalition government, the only case in which there can be a plausible argument for a mandate is if the coalition is formed before the election and the participating parties agree to campaign on a shared and well-publicized platform. Otherwise, voters cannot know with confidence exactly what catalogue of policy commitments they are endorsing with their votes.

My core argument, however, is that even when mandate majoritarianism is possible, there is little empirical evidence to support the existence of electoral mandates. Let me simply point to the assumptions about parties and voters that are implicit in this model and encourage each of you to ask yourselves whether they truly are characteristic of the parties and voters you know.

First, claims of election mandates require that the parties seeking such a mandate must explain to the voters during the campaign just what they would do if entrusted with control of the government. I reviewed the documents prepared by the Liberal and Labor parties in anticipation of the 2004 election, and I was genuinely impressed by the depth and breadth of information that both parties offered the voters, certainly in comparison with the U.S. party platforms that are adopted when we nominate our presidential candidates and then are immediately forgotten. If the Liberal Party documents that I examined were typical of all the others, voters in 2004 were presented with roughly 1500 pages of policy explication that presented the party’s intentions for the three-year period now ending. And since voters are required to choose among two or more parties, we need to add, at a minimum, the several hundred pages that comprised the 2004 ALP platform.

So we might say that Australia’s parties meet the first requirement for election mandates, or at least that they come closer to meeting it than political parties typically do. However, that in itself poses a dilemma. If the party’s manifesto is cast in more general terms, there can be all kinds of ways to write bills that arguably would implement the various items in it. So the voters might know in general terms what legislation to expect, but not the specifics. And, as we all know, in legislation as in so many other things, ‘the devil is in the details’ (a phrase which, incidentally, has been attributed to such luminaries as Michelangelo, Flaubert, and Mies van der Rohe, or perhaps it was Le Corbusier). But if, on the other hand, the party platform truly is so specific and detailed that it can be translated into legislative language without difficulty or ambiguity, then the governing party’s commitment to fulfilling its mandate leaves it little or no room to adapt to changing circumstances. No theory of electoral mandates can pass the proverbial ‘giggle test’ if it asserts that a party has the right and responsibility to implement its program of policy promises, but if it also contends at the same time that, of course, the party has to be able to make whatever changes in that program it considers necessary.

Second, mandate claims also must assume that the voters actually understand and evaluate the various parties’ plans and promises. The sheer size of the Australian party manifestos, with their supporting white papers and other documents, gives me absolutely unshakeable confidence that if comparable documents actually were presented by U.S. parties, very few of America’s voters would have more than the vaguest idea what is in them. Yet when the victorious party subsequently insists that one of its bills must be enacted and should not be changed, it can point to a paragraph or bulleted point in these documents as proof positive that a majority of voters must want the bill enacted because the party advocated it during the election campaign and the party won the election, from which it is supposed to follow that the electorate thereby endorsed that particular campaign promise. And if any government makes the same argument with respect to each of its campaign promises (although it would be impossible to enact all of them into law before the next election), it must assume that the electorate understood and supported each and every one of them. This assumption, too, is so implausible on its face that simply stating it suffices to refute it.

Third, the mandate theory assumes that voters base their election day decisions on their evaluations of the parties’ respective programs. In the United States, we know that this is not even remotely the case. Voters may prefer one party’s general approach to domestic and foreign policy to the other’s, but they also are influenced, and often more so, by such factors as their parents’ voting history, their own long-standing party loyalties, and, increasingly in the era of television, how much they like and trust individual candidates. We used to joke about ‘yeller dog Democrats’—voters who’d vote for a yellow dog so long as it was the Democratic party’s candidate. And American voters who claim that they don’t routinely vote for one party or the other proudly claim instead that they vote ‘for the man, not the party’—and, therefore, not the party platform. Voters whose choices do reflect their strongly-held policy views often are concerned intensely with one issue, whether it be abortion or gay marriage or gun control or immigration or the war in Iraq, and they pay much less attention, and give much less weight, to the others. Furthermore, voting in America often (perhaps typically) is retrospective; voters’ decisions are based on evaluations of how the party in power has performed since it took office, not on what it promises to do in the future. If most voters are not content with how the government has been performing, they are unlikely to be persuaded by its promises for the future. On the other hand, if most voters are satisfied with recent government performance, they are unlikely to throw the governing party out of office in favour of another party that can offer only assurances of its good intentions.

In short, the voting behaviour of Americans is entirely inconsistent with the assumptions and requirements of mandate theory. I venture to think that much the same may be true of the Australian electorate.

It seems to me, then, that on empirical grounds alone, mandate majoritarianism is a deeply flawed model of decision-making, even when the majority of the electorate has voted for the majority in parliament. It is much more difficult to defend the model when the parliamentary majority does not receive a majority of the popular votes—when the majority of votes go instead to the opposition and to other parties (and independent candidates) that are not in government. In Australia, for instance, neither Labor nor the Coalition often wins a majority of the votes cast in elections for the House of Representatives. In fact, in the 24 elections since 1949, only three times has the party winning control of the House and, therefore, the government, won a majority of the vote, and on two of those occasions it won by margins of 50.1 and 50.2 percent. What becomes of the argument that the government has both the right and the responsibility to have its legislative program enacted as it sees fit when the majority of the electorate, by that same theory, had rejected that program at the most recent election?  

Consensualism and its alternative

I believe the second model I wish to discuss, decision-making by consensus, is equally flawed, and for the same fundamental reason: it makes assumptions and imposes demands that, most of the time, simply are unrealistic.

In the United States, we frequently hear our elected officials say that we need to reach a consensus on how to address the pressing issue of the moment, whatever it may be. The goal should not be to enact the policy prescriptions of one party or political tendency—usually progressives versus conservatives—rather than those of the other (or others). Instead, the goal should be to bring together both or all parties in the legislature, and to bring together the legislature and the executive, in support of a policy decision that all recognise to be the right thing to do.

Let me share with you a few examples that I gleaned from the Congressional Record, Washington’s equivalent to Hansard, for January and February of this year, during the first days that the Democrats once again had majority control of both houses of Congress.

The Democrats in the House of Representatives flexed their new-found political muscle by passing a collection of bills during the first one hundred hours of session, the implication being that they were improving on the record of the House Republicans when they had taken power in 1995 and passed their collection of favoured bills during the first one hundred days. Both the content of the bills that the House passed in early January and the procedures by which they were considered were frequently contested, and bitterly so, by the House Republicans, who once again were learning the pain and frustration of being in the minority.

One such bill concerned the highly contentious issue of embryonic stem cell research which, for many, raises the spectre of abortion. During the debate, the new Republican leader in the House criticised the Democratic bill and the speed with which it was being propelled through the House by praising what he described as an alternative that ‘offers the potential for a new consensus approach’ to the issue. About a week later, at the conclusion of those first one hundred hours, one Democratic Representative was impelled to announce that ‘[w]e have set a tone for the 110th Congress that is one of cooperation, consensus, and compromise that extends beyond party lines.’ I recall no such announcements from the Republican side of the House chamber.

Soon thereafter, the new Democratic floor leader in the House of Representatives spoke in a debate about re-adopting a procedural rule that the Republicans had repealed. The rule had allowed, and now again allows, the delegates in the House who represent the District of Columbia and America’s other territorial possessions to cast some meaningless votes on the House floor. Referring to two of the most expensive U.S. government programs, Medicare, which provides health insurance for seniors, and Social Security, which provides income support primarily for seniors, the Majority Leader proclaimed that the ‘residents of the five territories should have a voice in shaping a bipartisan consensus that shores up the financial health of these vital programs.’ He said this, notwithstanding the fact that Social Security regularly is described as ‘the electrically-charges third rail of American politics’—to touch it is to risk almost certain political death—as well as the universal recognition that Democrats look for every possible opportunity to accuse Republicans of wanting to cut, gut, privatize, or otherwise attack the Social Security program.

Such paeans to consensus were not limited to the House of Representatives by any means. In February, the senior Republican Senator on the Finance Committee, with responsibility for reviewing and recommending bills affecting taxes, was discussing a provision of the income tax code known as the ‘alternative minimum tax.’ The distinguished Senator wanted to ‘remind people,’ he said, ‘that in 1999 we passed a repeal of the alternative minimum tax, but President Clinton vetoed it and we haven’t been able to repeal it since … .’ Moments later, though, he went on to assert that ‘[t]here is a bipartisan consensus that only complete repeal is an adequate solution to this problem’ of the alternative minimum tax.

Finally, what was the most contentious issue in American politics in early 2007, and now for that matter? The war in Iraq, of course. You may recall that the House of Representatives adopted a resolution expressing the opinion, without attempting to embody that opinion in law, that the President’s troop ‘surge’ was not a good idea. The Democrats in the Senate, with the support of a handful of Republican colleagues, attempted to bring a similar resolution to a vote, but they were stymied by a filibuster supported by most Republicans. One of the leaders of this debate was the former Republican chairman of the Senate’s Committee on Armed Services, who had broken with the President on this issue and who had been instrumental in drafting a resolution that the Democrats ultimately supported in opposition to sending more troops to Iraq.

During the debate, this Senator emphasised that he and his allies had no intention of promoting legislation that could in any way jeopardise the safety and well-being of American military personnel already in Iraq. He insisted that ‘[w]e solidly support that concept of no cut off of funds.’ ‘What do we do short of that?’ he continued. ‘Well, we have a debate. Somehow you have to have some focal point, something written down, some document in writing as to the ability of this institution, the Senate, to reach a consensus, and a bipartisan consensus, on how best we go forward with a new strategy in Iraq.’

There is one thing that all these references to consensus have in common. They all are nonsense.

I always have understood ‘consensus’ to refer to a meeting of the minds—a group of people all coming to a common understanding about something.[6] That agreement may be the result of a collective process of deliberation. Or it may be that each member of the group deliberates independently and then they come together to discover that they have reached the same conclusion. Whatever the process, there is implicit in the notion of consensus, to my mind at least, the idea that all members of the group, or at least the overwhelming majority of them, share the same understanding as to what is good, or what is right, or what is the best thing to do. Central to any consensus is, first, that it is supported by all, or almost all, of those involved, and, second, that they support it by choice, not because they are in any sense constrained or compelled to do so.

Conceiving of consensus in this way immediately reveals just how unlikely it is for us to expect to find a consensus on almost any issue of national significance that engages the attention of the Congress in Washington or the Parliament in Canberra. Political decision-making rarely is a process of politicians reasoning together until they all agree that there is a right answer to the question before them. In support of this contention, I need only refer to the thought of a distinguished but unrecognized American philosopher, my father, who used to say that when two people always agree, it’s certain that one of them isn’t thinking.

I will return to this theme shortly. In a practical sense, though, the line of argument I’ve pursued on this subject is irrelevant. Did that Republican Representative really believe that there was a consensus about stem cell research that was waiting to be revealed and embraced? Did that Democratic Representative really expect that his chamber would discover a consensus about how to address the exploding costs of Medicare and Social Security? Did that Republican Senator really believe that there was a consensus in favour of repealing the alternative minimum tax when recent and repeated attempts to repeal it had failed? And did his Republican colleague really expect that out of the Senate debate about Iraq would emerge a common understanding among Senators about what to do there? Did they really believe that consensus was likely, or even possible, in light of the different values that their colleagues hold most dear and the different interests, preferences, and needs of the people they represent? Of course not.

Instead, or so it seems to me, these references to consensus imply almost the opposite: that the problems under debate were so difficult, so divisive, and so intractable that, instead of confronting and addressing them as best they could, it was far easier for these legislators to talk about how wonderful it would be somehow to find solutions that everyone would prefer to the alternatives.

Let me take a momentary detour that, whether you believe it or not, will get me where I want to go. One effect and, to my mind, a benefit of electing MPs from individual constituencies is that each is linked to a particular geographical area and the people who live there. Consequently, it is possible to address MPs in debate not by name but by reference to the constituency—district or state—that each represents: not as ‘Mr. Jones,’ for example, but as ‘the Member for Buncombe’ (which, incidentally, is in North Carolina, and is the original source for ‘bunk’). This is the practice in the British House of Commons as well as the Australian House of Representatives, and, though sometimes honoured in the breach, the U.S. House of Representatives as well.[7]

Although this form of address sometimes sounds stilted and artificial to visitors, as do many of the formalities of parliamentary practice, it is explained and justified on the grounds that it de-personalises debate and reduces the level of animosity that otherwise might develop in the chamber. While that may be true, I think it also serves another related but distinguishable purpose.

It has been said that members of a durable parliamentary assembly need to believe and remember that the members of other parties may be opponents but they are not enemies. At the extreme, representative government is all too likely to collapse if members of parties or parliamentary groups believe that their personal well-being and security, and those of their supporters, are in jeopardy because they are in the minority. (Those who are convinced they are bringing democracy to Iraq might bear this in mind.) The concept of ‘enemy’ evokes images of war with victors and vanquished; the concept of ‘opponent’ evokes images of a game with winners and losers, but a game that will be played again and again so that today’s loser can hope to become tomorrow’s winner.

One reason parliamentarians find themselves opponents is because they have fundamentally different philosophies of government (or ideologies, if you prefer). What is the appropriate role of government in the society and economy, for example, and to what extent and for what purposes should the government intervene in the choices and behaviour of individuals as well as collective entities such as corporations? Another reason is that parliamentarians may have different understandings of how the world works. They may differ, for example, over whether international disagreements are amenable to negotiated resolutions or whether international actors respond only to the threat or application of force. Similarly, they may differ over whether helping the disadvantaged in society is best done by targeted government programs that, by definition, do not benefit everyone equally if at all, or whether economic growth and a favourable business climate is the surest way to promote prosperity in which all will share. On their most dispassionate days, it even may be possible for MPs to agree that they seek the same ends even if they have fundamental disagreements about the best means to achieve them.

There is a third reason why MPs disagree which is related to the other two but which is reflected in the impersonal and indirect way in which they often are expected to address each other. Under most party and electoral systems, including the form of proportional representation used in Senate elections, MPs represent geographic constituencies and those constituencies may have different needs and interests. MPs usually are expected to represent those needs and interests in the sense of speaking and advocating for them: ‘I rise to speak for Buncombe,’ as that Congressman is supposed to have said in 1820. Even in systems that elect all MPs from single national party lists, the parties may assign MPs of their party to develop strong ties with a particular community or region, perhaps the one in which each MP resides or was born. Generally speaking, we can expect this linkage between MP and a constituency to be minimal when national party organizations control the selection and re-selection of parliamentary candidates to the virtual exclusion of local influence, and when citizens base their voting decisions on national issues, without regard to how the various parties’ programs would affect their local areas, professions, or individual welfare.

If we accept a constituency linkage to be a typical characteristic of parliamentary life and work, we also must accept that MPs’ constituencies differ and that, very often, the differences among them are not differences of ideology or worldview; they are measurable differences. Some constituencies are richer than others; some have an atypical racial or ethnic composition; some depend more on industry and others more on agriculture; some rely more than others on exports for jobs and local prosperity. These possible differences could be multiplied. But the point is that, because of such differences, many proposed laws will benefit some constituencies more than others, and sometimes they even will benefit some constituencies at the expense of others. When one MP refers to a colleague who disagrees with him by referring to that colleague’s constituency, it is a way of reminding the MP that his colleague may be taking a contrary position in order to reflect and promote the real interests of his constituents.[8]

Examples abound. Free trade can benefit consumers by increasing the availability of lower-cost imports—clothing made in China or Bangladesh, for instance—but that competition can cost workers their jobs if the textile plants where they work cannot compete successfully. Farm price supports can help keep family farmers in business, but only at the cost of higher food prices and higher taxes. Increasing corporate taxes can reduce the need to increase individual income tax rates, but only at the cost of reduced corporate profits and, it is argued, reduced capital investment and shareholder value.  Drilling for oil in the Alaskan wilderness may eventually help control how much it costs me to heat my home and fuel my automobile. And so on.

It does not follow from examples such as these that politics is always a zero-sum game—that for every winner there is a loser. It does follow, though, that government policies and actions often have differential effects on different groups and regions and, therefore, on different constituencies. The challenge of law-making rarely lies in deciding who’s right and who’s wrong. Instead, the challenge usually takes the form of having to strike the most appropriate balance among competing needs and interests, even perhaps in debates over abortion: preserving the life of the mother versus protecting the life of the child. That is why a true consensus on important policy choices rarely is possible, because consensus implies a virtually unanimous agreement as to what is right, not what is the best we can do under the prevailing circumstances. And, returning to my earlier subject, that is also why I believe that mandate majoritarianism is undesirable on normative grounds, in addition to being unrealistic on empirical grounds.

Compromise as a virtue

When I was a boy, my schoolmates and I had to memorise the Preamble to the U.S. Constitution, which is so brief I’ll take a moment to read it to you:

We the people of the United States, in order to form a more perfect union, establish justice, insure domestic tranquillity, provide for the common defence, promote the general welfare, and secure the blessings of liberty to ourselves and our posterity, do ordain and establish this constitution for the United States of America.

These words had become so familiar to me that it was only years later that I stopped to think about what they mean, and about the act of faith that my constitution represents: that the particular set of political institutions it created somehow would lead to the achievement of all those goals laid out in the Preamble, such as establishing justice and providing for the common defence. The authors of Australia’s constitution saw no need for such a statement of goals, but I’m confident that the same assumption can be attributed to them: that the structure they were creating was more likely to promote the same future for Australia than any other on which they might have been able to agree.

It is on two of these goals that I want to concentrate for a moment, and those are the goals of promoting the general welfare and insuring domestic tranquillity, or social harmony as we might call it today. In brief, my argument to you is that majoritarianism based on claims of electoral mandates is not the most promising basis for promoting the general welfare or for insuring domestic tranquillity in the long run.

If by majoritarianism we mean only that law-making decisions are to be made by majority vote, then as I said earlier, we are almost defining a core element of what we mean by democratic governance.[9] Mandate majoritarianism is another matter altogether, because it transforms the power of the governing party (or coalition) to work its will in the parliament into both a right and a responsibility to do so. If that party is entrusted by the voters with the power of government, it is only so that it can implement the legislative program that the majority of the voters endorsed when they cast their ballots. Any concessions to the party or parties that are not in government would come at the expense of the government’s ability to fulfil the mandate it sought and received.

What then of the needs and interests of those constituencies whose representatives are not members of the governing party? Defenders of electoral mandates would argue, I suppose, that the victorious party already had taken those needs and interests into account in the process of formulating the program it presented to the voters. This is true to some extent, I’m sure. I would not want to imply that any responsible political party would deliberately ignore the needs and interests of any numerically significant segment of the population. However, I do question the notion that the leaders or members of any party are the ones best able to decide what should be done for or to those groups—whether regional, economic, social, or whatever—that largely opposed their party in the past and are likely to oppose it in the future.

It is an underlying tenet of democratic government that the people themselves are the ones best able to determine their needs and interests and what government actions will best serve them—what will promote their welfare as part of the general welfare. When it is impractical for us as citizens to speak for ourselves as our laws are being made, we rely on the people whom we have elected to speak for us. If all those who voted for non-government parties and independents believe that, once the election is over, they have no effective voice in government decisions until the next election rolls around, this situation cannot, in my judgment, effectively insure domestic tranquillity. On the contrary, long-term social harmony benefits from a generally-shared belief that the needs and interests of all segments of the population are being expressed forcefully in parliament and that their representatives are able to have a modicum of influence over the decisions made there.[10]

This finally brings me to the third of the three models of democratic decision-making to which I referred at the outset. If mandates are a myth, and a dangerous one at that, and if consensus is a chimera, then compromise is a virtue. The alternative to mandate majoritarianism is not ‘consensualism,’ it is a recognition that compromise is a good thing.

Here I am juxtaposing compromise against consensus. If the search for consensus is the search for what is right, the search for compromise is the search for what is best or most generally acceptable (which may or may not be the same thing) under the prevailing circumstances. Compromise requires the governing party to accept some limits on doing what it would like to do, and what it has the numbers and the formal constitutional authority to do, in order to take into account—legislatively, not just rhetorically—the needs and interests of those who voted against it. As my authority on this point, I will quote John Stuart Mill, who wrote that:

[o]ne of the most indispensable requisites in the practical conduct of politics, especially in the management of free institutions, is conciliation: a readiness to compromise; a willingness to concede something to opponents, and to shape good measures so as to be as little offensive as possible to persons of opposite views.[11]

I don’t mean for a moment to suggest that the minority, or the Opposition, should have as much influence over legislative decision-making as the majority, or the government. What I do mean to suggest is that the long-term interests of a nation are best served when the governing party has to ask itself what changes in its legislation it can accept that will ameliorate the detrimental effects that bill may have on certain constituencies, whether defined geographically or otherwise, without sacrificing the principles it has committed itself to promoting. And if the governing party is not inclined to view such compromises as desirable, which is what I would anticipate in the real world of politics, then it is good if the nation’s political institutions give some person or entity the power to make the government accept them as necessary.

Let me illustrate my contention with one concrete example. In the United States, there has been a recurring debate over whether Congress should increase the statutory minimum wage, the hourly wage that federal law requires employers to pay most of their employees. Most congressional Democrats have argued that the minimum wage has been too low for many families to support themselves adequately. Most congressional Republicans have argued that increasing the minimum wage would put too much pressure on many small businesses (which, they argue, are the engine that really drives job creation in the U.S.), forcing them to lay off employees or even close their doors. I’m not competent to say how many people would benefit from the wage increase or how many jobs or small businesses would be lost. But it does seem reasonable to me to believe that there is some truth on both sides of the argument. The pity is that it has been so difficult to agree on a compromise that hardly rises to the level of rocket science: coupling a minimum wage increase with tax breaks for small businesses; in effect, making the taxpayers pay for part of the cost of the minimum wage increase. Such a compromise—socializing part of the costs of a new or increased government benefit program, and thereby offsetting a focused benefit with a cost that is so widely distributed that it cannot really be felt by those paying for it—is an approach that often is available to government policy-makers.

Compromise by choice or necessity?

The last issue I want to address today is how such compromises are to be achieved.

In this context, let me first note that compromise seems to be taking on more of a pejorative connotation in Washington these days, sometimes being equated with the abandonment of principle. This reflects, I believe, the increased polarisation that has come to characterise American politics in recent years, as each of the two parties has become more homogeneous internally and as the policy disagreements between the two parties are thought to have widened. Party unity in congressional voting has increased as has the level of vituperation in political discourse. Compromise has become more difficult as the members of each party either have come to believe their own rhetoric about the other or they have become so trapped by their rhetoric that they are unable to justify compromises they otherwise might be willing to make. Furthermore, it is so much easier and perhaps even more satisfying for Representatives and Senators to use legislative debates as a way of appealing to their supporters in anticipation of the next election than it is to negotiate compromises with their political opponents from whom they have become increasingly estranged.

Although I have been away from Australia for four years now, it also seems to me that Australian politicians, at least those in government, often are reluctant to speak publicly about legislative compromise. The tenor of question time and the tone of each major party’s public comments about the other would be hard to reconcile with a visible approach to legislating that revealed the government to be taking the Opposition’s concerns seriously and accepting amendments that took them into account, while not satisfying them fully. As I documented in 2003, the voting record in the Senate does not reveal a government and Opposition always opposing each other; it is striking, in fact, how often the government and Opposition voted together on divisions during 1996–2001. That, however, is a hidden story of Australian politics. The kind of compromise approach to law-making that I am advocating involves a public recognition that compromise is taking place and that those compromises benefit the nation.

It is often said in my country, and I would not be surprised if it were equally true in Australia, that the time horizon of politicians is limited to the date of the next election. What may contribute to the long-term health of the polity (the general welfare) and its domestic tranquillity is going to be less important in practice to most elected officials than what will contribute to their shorter-term electoral success and their ability to promote their values and the interests they represent while they enjoy the power to do so. I acknowledge, therefore, that the winners of parliamentary elections are unlikely to concede voluntarily any of the fruits of their victories to the losers: the minority in the U.S., the Opposition in Australia. If so, the question then becomes whether there are conditions under which policy compromise is likely to be necessitated by the structure of political systems or the dynamics of electoral competition.

In a presidential system such as mine, the constitutional divide between the executive and the legislature can promote a competition for influence and an unwillingness of the Congress to accept presidential legislative proposals without change.[12] That is so even when the same party controls both branches of government. It is even more likely to be the case when the constitutional divide is exacerbated by divided party control, which has been more the norm than the exception in the U.S. during the last half-century.[13] When these factors are supplemented by relatively weak party discipline and an electoral system that encourages legislators to think first about how their electoral constituency, not their party leaders, want them to vote, policy compromises become almost inescapable, even when compromise is lamented as a necessary evil, not as a positive good.

The problem we have been having is the reluctance of the president to accept the inevitable need for compromise, and the temptation for members of the other party in Congress to insist on too much. As a result, we have had much too much talk of consensus and much too little willingness to engage in the search for practical compromise.

On the other hand, when both branches of government are controlled by the same, highly disciplined party whose unquestioned leader is the president who controls the political future of his party’s legislators, it would be surprising for him to sanction any voluntary concessions to other parties and the constituencies and interests that they represent. This has been the problem encountered in some quasi-, pseudo-, or proto-democracies, especially in Africa, where the dominant party has been largely the personal vehicle of its dominant leader, the president, and the legislature has thought of itself as a parliament, not a congress.

In a parliamentary system characterised by two dominant parties, there is little reason to expect a compromise approach to legislating, especially if each party is quite homogeneous and the two parties are quite polarised. One case in point has been Bangladesh, where there is raw hostility between the two parties, even though the policy differences between them are more difficult to pinpoint. When the leader of each party has accused the other of being complicit in murder, we should not expect them to relish the prospect of compromise decision-making.

In a multi-party system, which often is the product of proportional representation, we need to look at how and when governing coalitions are formed. Such a coalition can be the product of pre- or post-election negotiations among disciplined parties, negotiations that culminate in a specific and detailed political ‘treaty’ that binds the participating parties to a legislative agenda from which the dominant coalition partner must hesitate to diverge for fear of losing its parliamentary majority. If political parties are much weaker, on the other hand, and if coalition negotiations focus more on the allocation of portfolios and patronage, the dominant partner has more room for manoeuvre and more opportunity to make policy compromises if it chooses to do so.

It probably is fair to say that a compromise approach to legislating is most likely when parties are weakest and legislative majorities have to be assembled one vote at a time. I understand that, in the Kingdom of Jordan, for example, there are no parliamentary parties worthy of the name. MPs are elected as independents and do not then coalesce into unified and stable parties. This situation might seem to give the advantage to the government (putting aside the fact that the King really is the de facto policy-maker) because it is unlikely to face any organised opposition. However, I have heard it argued that the actual result is just the opposite, because the government does not have a parliamentary majority on which it can rely or that it can hope to mobilise. Instead, it must find the votes it needs where it can, and this situation allows each MP to bargain for whatever concessions he is able to extract.

The title of my paper speaks of mandates, consensus, compromise, and the Senate. What does this discussion of the first three mean for the fourth? I think the answer will be obvious to those of you who have followed my argument thus far.

In 2003, I argued that the Senate constituted the forum in which the Australian government can best be held accountable for its actions and decisions. The House of Representatives, controlled as it is by a highly disciplined government majority party whose members’ political futures depend largely on public support for their party’s government, have little practical incentive to question, probe, challenge, oversee, or investigate it in the public forums of Parliament. However much government party members might deny this, it is hard to deny the political logic that challenging their own party’s government in public can only damage the party and their own political prospects.

The Senate, on the other hand, has been in quite a different situation precisely because it has had a non-government majority for most of the last half-century. The combination of close party competition and the form of proportional representation that has been used in Senate elections since 1949 has put the government, whether Coalition or Labor, at a disadvantage in Senate elections, to the extent that the results of the 2004 Senate elections came as a surprise to many (including, I suspect, members of the Coalition government itself). Between 1996 and mid–2005, the non-government Senate majority had the institutional control, the constitutional power, and the political incentive to hold the government to account in a way that could not reasonably be expected of the House of Representatives. Although, in my judgment, the Senate did not live up to its full potential in this regard, it did contribute to the health of the regime by complementing the formality of government responsibility to the House with a more meaningful degree of accountability to the Senate.

Today I wish to argue that, for the same reasons, the Senate with a non-government majority has the same institutional control, constitutional power, and political incentive to compel the government to accept legislative compromises that it would be unlikely to make of its own volition. If the compromise approach I have described is preferable to the alternatives, and if the House of Representatives are very unlikely to be, and be seen to be, the forum for conciliation of which Mill spoke, then it is again to the Senate that we must turn. That is largely why I chose to look at the disposition of the Senate’s legislative amendments since John Howard became Prime Minister.

It would be unfair for me to speak at length here about findings that you do not have before you. So I will say only three things.

First, there is clear evidence of what we might call ‘compelled compromise.’ The government obviously has had to find additional votes in the Senate to pass its legislation, and certainly this has forced it to make some unwelcome compromises. However, I have not yet discerned a practical way to distinguish amendments made for this purpose from amendments that the government has proposed or accepted willingly in order to address weaknesses it had come to recognise in its own bills. Still, more often than not, government bills have survived Senate legislative consideration unscathed; and also more often than not, the Senate has not insisted on its amendments when the House has disagreed to them. My tentative conclusion is that this is a story of a glass that is half-empty, not half-full.

Second, I did not find much procedural evidence of legislative compromise after government bills have left the Senate and been returned to the House for further action. When the House has disagreed to a Senate amendment, the upshot usually has been either that the Senate has chosen not to insist on its amendment or the House eventually has chosen to accept it. There have been relatively few instances of the third possibility—the two houses agreeing on a presumed compromise in the form of an alternative to the Senate amendment to which both the Senate and the House of Representatives then agreed. It is unwise to ask a small body of data to support too heavy a load of inference. Even so, what I have found does suggest that what often is called the process of reconciling legislative differences between the two bodies has been less a process of reconciliation (that is, compromise) and more a process of allocating victories and losses between the Senate and the government acting through the House.

And third, July 1, 2005, did mark a turning point. After the government took control of the Senate, almost no non-government amendments were approved by the Senate, so there was no need for the government, acting through its House majority, to accept policy compromises. Virtually the only legislative amendments that Parliament has made since that date have been amendments proposed by the government itself. So has the change in party control in the Senate made a difference? I think so. Has that change contributed to the long-term health of the political system? I fear not.

Question — There’s a certain amount of literature which says that having total control is not only bad for the country but bad for the ruling party and the government. What’s your view on that?

Stanley Bach — The House of Representatives in Washington was controlled by the Democratic Party without interruption from 1955 through to the beginning of 1995 and I guess my answer is that Lord Atkin was on to something. It was clear that the Democrats in our House of Representatives came to take their power for granted and I think most observers would agree that in some respects they came to abuse it, in the sense of how they undermined minority participation and minority rights through various organisational and procedural changes they made. So I think some alternation in power will help to prevent that. We’ve had more regular alternations in power in the Senate and we have seen for this reason, among others, somewhat more restraint in a new majority trying to flex its muscles. I do have to say though, that when the House Republicans did become the majority in early 1995, at first they said: ‘We are not going to do to the Democrats in the House what they have been doing to us for the last 40 years’; and about two months later they started to say: ‘But all we’re doing to you Democrats is just what you did to us for the past 40 years.’ Now the Democrats are back in power, and I fear we’re seeing a process that social scientists call path-dependent. Once you start down some paths it’s very hard to turn back and go in the other direction because there are tactical advantages to some of these innovations. It’s a destructive spiral.

Question — But what about being able to enact your program by having total control in both houses plus the executive government? You’re then tempted to enact some of the more bizarre features of your program and this gets you into trouble.

Stanley Bach — This is a question that’s a little bit more difficult for an American to respond to, because you’re presuming our parties have programs. Our parties are considerably more polarised now than they were before the disappearance of the conservative Democratic south and the disappearance of the moderate, not to say liberal, Republican north-east. When you have both Houses of Congress controlled by the same party, about the only real power that the minority party has is its ability to filibuster in the Senate. You see, in the Senate you can’t vote to pass a bill if there are senators who still want to talk about it, no matter how long the debate goes on, unless you can assemble a three-fifths vote to stop the debate, and the majority party in the Senate rarely if ever controls three-fifths of the seats. So a determined minority in the Senate can generally bring the institution to a standstill. Before this most recent election, the Congress was completely under Republican control. The Democrats in the Senate started to elevate the filibuster into a matter of high constitutional principle, that it is part of the system of checks and balances, because it was the only effective power lever that the national Democratic Party had. This is what I sometimes call a constitutionalism of convenience, just as it will be interesting to see what the ALP and the Liberals have to say about the Senate if the situation next July should be reversed.

Question — I’m sympathetic to the general call you have for compromise, but I wish to call your attention to the fact that one of the reasons why we had a majority Liberal Senate in 2004 was because of the demise of the Australian Democrats, who were exactly the kind of compromising party that you call attention to. Unfortunately, they compromised once too much by being willing to pass the GST, the sales tax that almost got the Howard Government beaten in 1998. It seems as though a compromise is getting it sort of in the neck, and I would like you to comment on that.

Stanley Bach — I’ll make two comments. First, politics is a messy business. Second, I’d be a little bit careful about generalising from a sample size of one. Also, I do take your point about the demise of the Democrats.

Question — I refer to your candid comments in the beginning that a journalist and a presenter on the radio did not read the policies in the last election. If a journalist doesn’t do that, what chance does a member of society have? Also, can you think of a way that you can make things easier on the general public in terms of information, and do you think the internet has a part to play in this?

Stanley Bach — Well the internet certainly has a part to play in making information available. That’s the resource to which I turned when I wanted to confirm my supposition that for the 2004 election Labor and Liberals had put out massive policy documents. So you can make the material available, you can make it readily accessible, but people have to choose to find it and they have to choose to read it. But you have a dilemma: the longer the documents are, the fewer the number of people who are going to read them; but the shorter and more general those documents are, the less value they have for predicting what parties in government will do, and the greater the gap between some general statement like restraining government spending or cutting taxes, and a specific bill that’s to be introduced in parliament, where we’re told: ‘This is a bill to implement that promise.’. So no, I don’t have a good solution.

Question — I have a question that’s also related to the Democrats. The Democrats tried to carve out a position where they said they were there to keep the bastards honest. But they then seemed to adopt a policy of keeping the bastards dishonest. What would happen is that the government of the day would sit down and produce a policy document, take it to the electorate, and then we’d have an election around it. We had a big election around GST, and then they won the election and then they wanted to implement their policy that had been fully costed in great detail. The Democrats turned around and said: ‘We’re there to keep you honest. But we’re not going to let you introduce your reform.’ This has happened again and again. For example, the superannuation surcharge. The government turned around and said: ‘If elected we will wind back a tax.’ The Democrats turned around and said: ‘Well, we’ll keep you honest. You’ve won the election, but we won’t let you do what you promised to do.’

Another model might be to sit down and espouse a balance of power party that says: ‘We will require a government to enact the legislation, we’re not going to impede it, but we’re also going to sit down and say we’ll agree to allow you to enact your proposed reforms, but we’ll tie it into making sure you enact all your promises and not just cherry-pick the ones that suit you’. So there is another model that you could come up with where governments which have a mandate are actually given the opportunity to enact it, but also requires that they enact all their mandates and not just cherry-pick the ones that suit them. I think that’s why the Democrats fell down, not because they compromised per se, but because they had this duplicity in their position.

Stanley Bach — The GST was a matter of some controversy wasn’t it? That occurred before my first visit here, so I don’t claim to know very much about it, but as you may assume from what I said earlier, I don’t have a great deal of sympathy with your argument because I don’t think that government can make an empirically valid claim that it does have a mandate to enact everything in its program. Before I could accept that kind of approach, I’d want to see some opinion data, for example, that gives us some notion of what the Australian electorate actually did know when they cast their votes and to what extent they cast their votes on the basis of that information. Earlier this week, I went to a seminar at the ANU where Professor Ian McAlister presented some results on Australian public opinion derived from election surveys from 1987 to 2004. I suggest you get a copy of this from the ANU, or take me on faith when I say that among all of these survey results there is very little that would give any support to folks who would claim there really is an election mandate out there.

Question — You referred to a tendency towards an increased presidential style of government in Australia, and certainly that same comment has been made in respect to the United States. I refer particularly to the work of Chalmers Johnson, where he asserts that what we have in the United States is increasingly an imperial presidency, which implies a breakdown in the whole range of checks and balances that are supposed to characterise a republican form of government as supposedly set out by the founding fathers. An important element of this argument though, is that he also talks about a military industrial complex, or rather a military industrial Congressional complex, in which, on the basis of looking after their constituents as you put it, together with the fact that the military industrial complex has plants and therefore employment in every state, the Congress is really unable to put pressure on the executive on the matter of foreign policy. Most of what you’ve said has been in terms of very good comments about procedures, but what this argument is saying in respect to the United States, and possibly also Australia in a much smaller way, is that the breakdown of the system of checks and balances and the rise of an imperial presidency has a serious effect on policy content, particularly in relation to foreign affairs or international politics. This is a very difficult issue, a very complex issue but I was wondering if you have a comment on that.

Stanley Bach — You have opened not one can of worms but quite a few. I’m aware of Johnson’s work. I haven’t read this book, but I know of it. Several perhaps disjointed responses: first your reference to the imperial presidency. Arthur Schlesinger wrote a book entitled The Imperial Presidency in 1973. We’ve been there before. Your reference to the military industrial complex, as I’m sure you know, evokes memories of President Eisenhower in the late 50s. So the kinds of concerns that you’ve expressed either as your own or reflecting Johnson’s work are in most part not new; concerns about an imperial presidency are very much tied to the prominence of foreign policy in the debates of the moment. Nixon’s imperial presidency was all in the context of winding down the Vietnam War, and I think with respect to Bush in the context largely of the Iraq War.

It is true that there historically has been a considerable reticence on the part of members of my Congress to challenge presidents on questions of foreign policy, both because lives are at stake and because these are issues frankly that most of our representatives and senators don’t feel nearly as competent to address as questions of domestic policy. So yes, there does tend to be much greater deference to presidents because it’s much easier for members of Congress to stand back and if things go sour to say: ‘Well, you won’t find my fingerprints on it.’ We did have resolutions with regard to sending troops into Iraq, and those of you who follow American politics may know how some Democratic senators and presidential candidates have had to tiptoe and tap dance around explaining why they voted the way they did.

Johnson’s argument, as he summarises, teeters on the edge of conspiracy theory; that there is this military industrial complex that really is controlling decision-making. I think that’s much too much of a simplification. It’s true that much spending on defence turns into spending that occurs within the United States on military procurement, and that is a wonderful boon for the states and districts where the military plants are located, but what the argument doesn’t take into account is instances in which what you have in Congress is competition for those contracts. I remember one case, I forget how many years ago it was, when Congress was about to provide money for some aircraft I think it was, and there was a heated debate between the senior senator from Washington and the senior senator from Georgia about the merits of two different models for this plane, and it wasn’t a coincidence that Boeing is located in Washington and Martin Marietta is located in Georgia. So it’s not any kind of monolith controlling government decisions.

Question — My question relates to the difference between consensus and compromise because I think you’ve put far too high a bar on consensus. The idea that people will always be in harmony is of course nonsense, but it seems to me the difference is that the American system assumes that you start with your position; you have an interest group, their position is fixed, and then it gets traded off in a complex series of manoeuvres through Congress and so on. Consensus it seems to me is best typified by systems of proportional representation in lower houses, and assumes that while people come from different places, there is movement in the process of debate, and they end up with a position that they could at least all live with, if not wholeheartedly support. So it seems to me that there is a conceptual difference there between compromise and consensus that you didn’t allow for.

Stanley Bach — Where you have proportional representation you don’t have powerful interest groups, is that right?

Question — You do but it plays out differently and they perhaps end up by being a little more muted than can happen in America for example.

Stanley Bach — I think there’s some truth to what you say but I think to an extent it’s a caricature about how political decisions are in fact made in the United States. Yes, interest groups are influential, but they are not the ones who really make the decisions. After all, bear in mind, what interest groups have to offer are two things: information and money. Now the information is valuable, but if members of Congress receive information from lobbyists and they discover the information is false, they are not going to listen to those lobbyists again. So lobbyists have a real incentive to tell the truth. They just don’t tell all the truth. They tell the part of the truth that supports their position. Well, that’s why you need a non-partisan and professional research staff, to always say: ‘On the other hand’. The other thing that interest groups have to offer is money. Money is used for campaigns. What’s the purpose of campaigns? It’s not to raise money, it’s to attract votes. So an interest group is only going to have an effect of one of our representatives or senators to the extent that the funds they provide can be used to attract votes, and the senator is not going to say to the interest group: ‘You give me a campaign contribution and I’ll vote your way’ if he know that’s going to alienate a significant share of his constituents, because it’s just not worth the price. What he is ultimately most interested in is not the money, it’s the votes. So that’s one general kind of reaction.

With regard to proportional representation, the argument you’re making is one that’s been made elsewhere. I don’t pretend to be an expert on PR; I’ve never lived under a PR system, but for those of you who are interested I commend to you the work of Aaron Lijphart from the University of California San Diego, who has written books on patterns of democracy in which he makes just this kind of argument. It’s complicated because it depends not just on whether you have PR but what kind of proportional representation and what kind of party system. This is why I said at one point that if you’ve got PR with many parties and none of them approach a majority, you could either have a treaty signed between parties before the election, maybe a bit like the Liberals and the Nationals, or a voter votes for a party without any real idea of whether that party is going to end up in government and if so, who its allies would be. A classic case was for some years in Germany where you had the CDU and the SPD representing generally the centre right and the centre left, and in the middle you had the Free Democrats, who really did usually hold the balance of power. But if you were voting for the Free Democrats very often you didn’t know whether you were voting for a coalition with the SPD or the CDU. So I sympathise with the arguments for PR, but first I want to know what kind of PR and I want to know something more about the political and electoral system in which it’s going to be imbedded. It’s very difficult to predict exactly how it will work.


*     This paper was presented as a lecture in the Senate Occasional Lecture Series at Parliament House, Canberra on 19 October 2007.

[1]     Department of the Senate, 2003.

[2]     ‘Senate Majorities and Legislative Outcomes: the Fate of Senate Amendments, 1996–2006,’ October 2007.

[3]     Business of the Senate is published annually by the Department of the Senate. Statistics contained in it are available online at

[4]     The potential effects of those elections are not limited, of course, to the Senate’s influence on legislative decisions. However important the Senate’s legislative responsibilities are, it is equally important for the Senate to hold the government accountable for how it has, or has not, implemented existing laws. In this respect Harry Evans has reviewed the first year of government control of the Senate and found that ‘[t]he government majority in the Senate has greatly increased the ability of the government to do what it likes and not to explain itself except to the extent it chooses.’ Ch. 10, ‘The Senate’ in C. Hamilton and S. Madison (eds), Silencing Dissent. Allen and Unwin, Crow’s Nest, NSW, 2007, pp. 220–221. Specifically, he notes such developments as reduced time for Senate committee review of government bills, regular defeat of motions ordering the government to produce documents, and, perhaps most depressing of all, fewer sitting days for a body that already was not exhausting itself with the number of days on which it worked. Since then, the Senate also effectively abolished the references committees that had been chaired by non-government senators.

      It should be noted that, in recent years at least, the government was not always responsive to Senate attempts to obtain information about its decisions and activities, even when non-government senators were in the majority. To this outside observer, in fact, the government’s attitudes toward these efforts sometimes seemed to be dismissive or even contemptuous. But to the extent that these attitudes did prevail then, and may prevail now, it must be in part because the Senate has been reluctant to assert vigorously its full range of constitutional powers and to insist on due respect for its constitutional responsibilities and prerogatives. If the government has ‘dissed’ the Senate, as young people in America now are prone to say, the Senate itself must accept some of the responsibility by allowing it to happen.

[5]     Elaine Thompson, ‘Australian Parliamentary Democracy After a Century: What Gains, What Losses?’ Parliamentary Library at

[6]     In fact, the dictionary on which I’ve happily relied for more than four decades offers ‘unanimity’ as its chosen synonym for ‘consensus.’

[7]     One reason the practice is not always followed in the last of these is that referring to the ‘Gentleman from California,’ for instance, is not very helpful when there are quite a few men representing California districts. I suppose that is why Australian Senators are identified by name; with 12 Senators from each state, referring to the ‘Senator from Tasmania’ would not be very discriminating.

[8]     The larger the constituency, the more diverse it is likely to be; and the more diverse constituencies are, the more they will tend to be like each other in their needs and interests. Sometimes what is most important, however, is not the nature of the constituency as a whole, but the nature of the winning MP’s electoral constituency—the majority that has elected him or her to office. In the U.S. House of Representatives, it is not unknown for a very progressive member to be succeeded by a very conservative one, or vice versa, because the district is fairly evenly divided politically and election outcomes are decided by the swing voters in the middle who are most likely to vote differently from one election to the next. The conservative MP will perceive the needs and interests of the constituency quite differently from his or her progressive predecessor because each has a different mental picture of it in mind, no matter how an outside observer might view it.

[9]     Although the size of the required majority can be an issue. As I observed earlier, constitutions and parliamentary standing orders may require absolute or larger majorities, such as two-thirds votes, for certain purposes, and it can be said today that the U.S. Senate effectively requires a three-fifths vote to bring any contentious proposition to a vote.

[10]    Even if the government party could and did take all needs and interests into account as it formulated its election manifesto, that would not suffice, because it is important for all segments of the population to see themselves as having some effect on government policy-making.

[11]    Quoted by David Hamer in his Can Responsible Government Survive in Australia? 2nd Edition. Department of the Senate, Canberra, 2004, p. 347.

[12]    During times of crisis, the U.S. Congress sometimes has set aside its capacity for independent judgment, but not for too long. Examples are the Great Depression of the 1930s, when Congress rubber-stamped the first legislative initiatives of the Roosevelt Administration, but then had second thoughts about the risk of excessive government intervention into the economy; and the national security crisis following September 11th, when Congress approved executive powers that, upon reconsideration several years later, have raised serious questions about excessive government intrusions into personal privacy and civil liberties.

[13]    When Senate Republicans were in the majority in Washington before the 2006 election, they became frustrated by Democratic filibusters that prevented votes to confirm some of President Bush’s judicial nominees. Republicans discussed exercising what came to be called the ‘nuclear option’ that would effectively have changed Senate rules to undermine or prevent at least some filibusters. In response, some Senate Democrats spoke of the right to filibuster on the Senate floor as if it were an integral part of the constitutional scheme of checks and balances. It was implied that Congress can check and balance presidential power only when the party opposing the president has the ability to block the president’s legislative initiatives (and nominations) in at least one house of Congress. Such an argument has no basis in constitutional theory or constitutional law, but I admit that it did, and does, have some practical political resonance.

Prev | Contents | Next

Back to top

Facebook LinkedIn Twitter Add | Email Print
Back to top