Find us on Google+

Saturday, 28 July 2007

Buying the electorate…

A new IMF Working Paper by Morgan and Vardy draws the following conclusion On the Buyabilitiy of Voting Bodies:

We found that increasing the size of the body increases the cost of successful vote buying when there is only a single interest group seeking to influence the outcome. In contrast, when there are competing interest groups, larger voting bodies may actually be cheaper to buy than smaller voting bodies.

It is often said that the cure for vote buying is to expand the size of the electorate. If you double the eligible voters you can raise the cost of successful corruption. This intuition relies on the direct effect that such an expansion of the electorate has on the costs of a single, monoposonistic lobbying party. While the size of the bribes remains the same, the lobbying party will have to bribe a larger majority of the electorate than before to achieve the same outcome. What Morgan and Vardy seem to demonstrate is that competitive political processes like multi-party elections, introduces a “strategic” reason for bribing voters that is absent in non competitive process (e.g. one party elections) – making that direct effect less potent.

For us in Zambia, the lessons might be two fold looking towards 2011: a) simply increasing the number of registered voters is not a guarantee that the outcome is more representative of the people’s true wishes; and b) increasing competition among political parties does not always mean that the voters wishes are reflected in the political processes - more electoral competition may just mean more corruption as they seek to out do each other!

To really make progress against electoral corruption, we should strive to develop policies that strength the institutions that oversee the electroral process. Considerable effort should be spent on taking forward some of the recommendations of the
Mungomba Draft Constitution (para 112 – 119) on structural reform of the ECZ. Essentially the ECZ should be independent of the Executive and appointed by the Appointments Committee, which is appointed by the National Assembly.

8 comments:

  1. Cho,

    I realise that you did not write the article cited, nor am I actually opposed to it's apparent conclusion that local governments may be more resistant to corruption relative to their larger national cousins than previously thought.

    However, when I encounter in an introduction phrases such as, "We show that, regardless of competition, the option to contract on both votes and outcomes is worthless, as it does not affect buyability as compared to contracting only on votes. In contrast, when interest groups can contract on votes and vote shares, we show that voting bodies are uniquely at risk of being bought", then I start hunting for definition of terms, because while each word is familiar, the way they are being put together is unfamiliar. Adding to my frustration, unless I can track down and evaluate the model concocted by, "Groseclose and Snyder (GS, 1996)", then I cannot hope to understand the way that data is being interpreted in this study.

    What is the distinction between a "vote" and a "vote share"? How does one "contract" on a "vote" and/or an "outcome"? If I am one of the majority of major corporations who give money to both candidates in any pivotal American election, am I aiming at "votes", "vote shares", or "outcomes"? As for discriminatory vs. non-discriminatory "vote buying", is there even an historical case of the latter within the last two centuries? These days even mass-media advert buys are discriminatory, or else fire the campaign manager.

    It does not take a political veteran to recognize that contentious legislative issues can hinge upon the opinion of a single powerful individual, such as a committee chair or a ministry appointment, or just an MP who finds themselves the only one uncommitted and in possession of the swing vote, and that therefore in any circumstance where more than one opinion shapes the form of government interaction with the public, capture of a department relevant to any single issue has political value. In as much as the size of a voting body necessarily increases the potential number of subgroupings within it (given the immutability of an individual human as the basis unit), then the potential political value of capture of any one grouping within it is both easier and less likely to pertain to the whole of the political capture equation. Thus given more people, it takes more actors (and/or more money) to reliably identify the groupings which translate into formal political action at the polls.

    "Absent competition, increasing the size of the voting body provides effective protection against vote buying. In the presence of competition, this is no longer true. In this case, larger voting bodies may be more buyable than smaller voting bodies." Absent competition one can do a lot of things which go counter to common sense or the common good. In the presence of competition, that which leads to electoral victory will be discovered and employed, every time. These are also propositions which "real world" evidence will easily back up. [Also if anyone can tell me what this nearby bit means, "In this paper, we are also interested in circumstances where (aggregate) outcomes are contractible. This makes pivotality important.", it would be appreciated.]

    Couldn't these results be phrased as, "larger polities are more expensive to buy, but because the payoff of success is so much greater than with smaller polities, more people will try and thus those who succeed will appear to be more efficient than those whose success comes on the less contested local electoral 'battlefield'"?

    The literature on the subject of strategic voting by individuals is spotty at best, so I applaud the authors' intent, however I suspect that their basis in undertaking the study is far too narrow, and that the "supply side" nature of the analysis misses the mark. I would recommend, Making Votes Count, by Gary W. Cox as a good starting source for the interested reader, unproven hypothesis, but lots of international data imbedded throughout.

    [One does have to appreciate however one apologetic footnote, "We will sometimes refer to vote buying contracts as 'bribes.'. This is for succinctness only and not an expression of the legality (or lack thereof) of a particular contract." One is forced to wonder how exactly how the authors came upon their "real world evidence" in such context. But my personal favorite is the marveling at how the limitations of the model itself make the numbers move: "Figure 3 displays A's total costs for this example, as the size of the voting body varies. It is interesting to note the points in the figure where the costs jump. These jumps occur when the number of moderates increases by exactly one voter— which happens when the size of the voting body increases by 10 voters— until there are three moderates, which occurs when the voting body consists of 21 voters."]

    ReplyDelete
  2. Yakima,

    ”I realise that you did not write the article cited, nor am I actually opposed to it's apparent conclusion that local governments may be more resistant to corruption relative to their larger national cousins than previously thought.”

    I am relieved!
    I was worried I would have to explain it :)

    ” Adding to my frustration, unless I can track down and evaluate the model concocted by, "Groseclose and Snyder (GS, 1996)", then I cannot hope to understand the way that data is being interpreted in this study.”

    It’s one of those papers written for the political science audience and is not available electronically I think. I have e-mailed the authors to track it down.
    Failing that I hope to get a copy via British Library – who should have an electronic version.

    ”What is the distinction between a "vote" and a "vote share"? How does one "contract" on a "vote" and/or an "outcome"? If I am one of the majority of major corporations who give money to both candidates in any pivotal American election, am I aiming at "votes", "vote shares", or "outcomes"?”

    My limited understanding of this complex area suggests that contracting on outcomes is when the pay out of the bribe is linked to the result. So the example the author gives is the Salt Lake Olympic Committee scandal where the people not only got bribed for voting, but also got a bonus “bribe” if the bribe succeeded in their favour. That actually is an example of contracting on the vote AND the outcome – the outcome being the “result”. I guess you could contract on outcome purely by giving ex-post bribes – but that is obviously not a credible commitment since you can renege.


    ” [Also if anyone can tell me what this nearby bit means, "In this paper, we are also interested in circumstances where (aggregate) outcomes are contractible. This makes pivotality important.", it would be appreciated.]”

    The paper is not well explained I guess!
    I have tried finding a simple explanation of “pivotality” – which is really a “median voter” used in rational choice based models (e.g. Downs Model). Hope this explanation is clear:

    Pivotality is the calculated probability of casting a deciding vote that enables a preferred party to win and prevent a less preferred party or parties from doing so. With regard to turnout this can be simplified as the calculation of whether casting his or her vote will make a difference. If voting is unlikely to achieve anything or result in any benefits to the voter, why should he or she vote? The voter is interested, according to rational choice theory, in making a decision based on ‘utility’ and therefore seeks to determine which party has implemented or proposed policies that will be of perceived benefit to him or her. Pivotality interacts with benefits as any benefits are discounted by the probability that that an individual can exert a crucial or pivotal effect on the outcome. The voter must also assess the costs associated with voting; the time needed to vote and to acquire the information to make an informed choice.
    Source: http://www.epop06.com/papers/PhelpsEPOP%202006.doc


    I guess in the quote you cite from the IMF paper, pivotality becomes important because it is the ‘pivotal voter’ that really sways the outcome. So an efficient contract should be seeking to capture that ‘pivotal voter’. [At least that’s my understanding!]


    ”I would recommend, Making Votes Count, by Gary W. Cox as a good starting source for the interested reader, unproven hypothesis, but lots of international data imbedded throughout. “

    I hope it is a very introductory text without any pivoting!! I look for it. I have never really read around voting beyond the basic Downs model, but the literature seems very rich – with fascinating issues.

    ReplyDelete
  3. Cho,

    Thanks for the clarifications, especially on "pivotality" (which could as easily have referred to the ease with which a voter can be persuaded to turn on an ideological "axis", or the degree to which their own movement is amplified into movement of their entire "wing"). The Salt Lake Olympic example of "contracting on both votes and outcomes" is interesting, especially given the conclusion that such bargains are "worthless" in that they produce the same results as "contracting only on votes" (but with presumably higher costs?).

    One thing which does seem potentially useful about a scientific determination of the ideal focus and distribution of political bribes is that regulators could also use the data to focus their own oversight efforts. If outcome contracting is inefficient, then assume that the most efficient (and presumably successful thereby) bribe payers wouldn't waste their money and energy on such things, so watch the highly pivotal voters closely before they vote.

    Only then Heisenberg rears up, and the observer changes the thing observed, and if outcome based contracts become more successful relative to vote based contracts due to the activities of regulators, then bribe payers change their habits and a whole new observation is required to assess the changes.

    My guess is that, at least for some of the more influential american politicians of late, those with high "pivotality" quotients on key issues, long term relationships with powerful lobbies has encouraged outcome based contracting. The prevailing paradigm among retiring congresspeople and executive appointees is to accept a highly paid position from one or more of the companies or political action groups they had been regulating while in office. So the author of the "Medicare Prescription Drug Act", which forbids competitive bargaining for lower drug prices, retired his seat at the next election and accepted a $2m/year (plus benefits) position from a leading drug manufacturer. The appointed federal regulator who signed off on an agreement to provide central California water priority to a few large farmers over municipalities for the next 25 years, is now the chief lobbyist for the same group of farmers, etc. and so forth.

    The outcome based contract enables "plausible deniability" that the two items, regulation and subsequent employment, are causally connected. In fairly obvious cases we can assume that regulatory favoritism influenced the subsequent employment compensation packages received, but proving it is something else entirely. After all, the companies have been making record profits off their government contracts, and can easily afford generous salaries to qualified employees, and who more qualified for an administrative position in the industry than the guy who wrote the regulations to begin with? Plenty enough there to squeeze some "reasonable doubt" into the legal equation.

    The book by Cox is pertinent to the Zambian context in light of the electoral process changes suggested by the draft constitution. It provides analysis of the effects of differences in voting structure on contestability of elections and consequent "strategic voting" habits of electorates. Unfortunately my copy is on loan to group looking to reform the local electoral process, so I can't reference anything verbatim. I am especially fond of the appendices, which summarize the peculiarities and electoral differences between hundreds of representative democratic nations.

    ReplyDelete
  4. Yakima,

    As promised.
    The G& S paper can be accessed here.

    Let me know if you encounter any difficulties!!

    ReplyDelete
  5. Cho,

    Thanks much for tracking that paper down, the model is interesting. I was particularly struck by this paragraph from footnote 4:

    "We allow v, to be infinite. That is, legislators can have such an intense preference for x or s that no bribe of any size will make them change their vote. Of course, if a majority of the legislators have such intense preferences, then no vote buying takes place, and the model becomes trivial."

    The implication is that the most reliable way to reduce bribery of legislators is to elect politicians with intense policy preferences rather than those advocating compromise or campaigning on their professed flexibility. In such a case, a culture of competitive bribery may in fact be preferable to the alternative of a highly polarized political landscape.

    ReplyDelete
  6. "In such a case, a culture of competitive bribery may in fact be preferable to the alternative of a highly polarized political landscape."

    Intresting conclusion. Surely it is difficult to express a preference between the two. Isn't the key whether the legislators (lets call them MPs for simplicity) preferences are aligned with the population they represent?

    ReplyDelete
  7. "Key?", yes, most certainly, however assigning a normative label like "good" or "bad" to the effects of closely aligned preferences between voters and MPs in an atmosphere which discourages compromise is more difficult. Optimistically we may hope that while individual MPs remain unbribable in their defense of shared voter preference, they will remain equally committed to restraint when it comes to obstructionist or violently inciteful tactics to achieve their political goals or retain their office.

    The degree to which a given population is likely to be truly aligned in strongly held opinion with their MPs depends largely on the electoral structure of the national government. The current system of winner-takes all, regional geographic district elections tends to produce relatively low percentages of aligned voters. [Apparently it is based on the old and unpopular US model. If I had my copy of Cox, I could give an exact number of countries which use it, but I am pretty sure I could count them on one hand.]

    In closely contested districts with such systems, 51% of votes is not uncommon for victors, less where simple plurality is sufficient rather than outright majority. Even if turnout at the polls is relatively high, let us suppose 75%, then the number of eligible voters who actually voted for the victor drops to 38.25%. If we add to this the number of voters who based their vote on party loyalty, then the degree to which the MP differs from the party line will also reduce the number aligned with them (this also presumably becomes more likely the more intense the MPs preferences are). Other voters find that they don't really agree with any of the choices they are given, and make their decision based on a "lesser of evils" comparison. This in turn encourages a contraction in the number of parties contending for a given seat, ironic for a group of voters dissatisfied with all candidates.

    The proposed changes in the draft constitution which would allow for proportional representation by party list for a substantial number of seats (I believe the current proposed figure is 40%) would alleviate this problem somewhat. Since voters in the minority within their district will influence the distribution of seats off the party lists, they will be more likely to cast their vote for an aligned MP rather than an unaligned one with a greater chance of victory at the district level. Improved local governance as well as a strong House of Chiefs with alternative electoral procedures would also improve the odds of a given voter being aligned with one or more of their legislative representatives.

    Theoretical models which make heavy use of ICT to provide near-perfect alignment are currently beyond the Zambian capacity (and likely any nation's current capability) as discussed in the various Zamtel/Mulongotism blogs.

    ReplyDelete
  8. Yakima,

    Interesting and thoughtful reflection as always.

    I had to think through this one over and over again :)
    I think there are four separate questions here which are worth drawing out to aid the discussion:

    1. Does realignment matter or is it distortionary?

    2. If the answer to 1 is "yes", does it matter once you have aligned the voters with the MPs, whether that leads to less compromise or not?

    3. Is it feasible to successfully align voters preferences with MPs own preferences?

    4. How far does the current Mungomba draft proposals take us towards that realignment, if indeed it is desirable to do so (i.e. if the answer to question 1 is yes)?


    On (1), I think we probably agree that re-alignment matters. After all it is the only reason why people bother to vote. Voters vote for MPs they believe are closer to their position, even in the face of bribery. If you tell people your realignment does not matter, they wont bother to vote.

    On (2), my position is that once the realignment takes place "perfectly" its irrelevant whether compromise takes place or not. In fact the issue of compromise does make the drive for realignment less important. In my view voters already 'internalise' the costs of not being able to reach a compromise before they express their preferences. Unless we believe that to be wrong, I see no immediate reason why the strive for realignment need to consider the costs of failing to reach a compromise.

    On (3), I agree with you that perfect voter alignment is near impossible. But I think that is the same as simply saying there's no such thing as perfect information. Which is isn't actually a problem as long as we can make sure that when MPs vote, they feel the costs of their actions. As Landsburg puts it best : "the problem with democracy is not politicians kowtow to financiers and lobbyists; it that politicians kowtow to their own constituents. In other words, the problem is that politicians have little incentive to consider the costs of their actions". I read this as saying, voter alignment has some negative costs [turning my answer to (2) upside down:)] but what matters is the broader problem that electoral systems provides no incentives for politicians to look beyond their borders. It turns Landsburg propopose two voting opportunities for each citizen one for where they live, and another for the place or their choice - or alternative, allocate constituents alphabetically. So people with names beginning with AB are part of one constituent, people with AC are part of another and so forth.

    On (4) - I'll come back with more thoughts :)

    ReplyDelete

All contributors should follow the basic principles of a productive dialogue: communicate their perspective, ask, comment, respond,and share information and knowledge, but do all this with a positive approach.

This is a friendly website. However, if you feel compelled to comment 'anonymously', you are strongly encouraged to state your location / adopt a unique nick name so that other commentators/readers do not confuse your comments with other individuals also commenting anonymously.