Monthly Archives: March 2015

Paying politicians

My friend, Simon Hix (LSE) retweeted a column in the Economist about the recent Straw-Rifkind cash-for-lobbying scandal.  The Economist’s view was that these sorts of issues (broadly, MPs having to seek outside income) could be avoided by paying UK MPs more. I’ve actually been doing research on this topic, and I’ve read a fair number of academic papers on the topic.  Let me offer some off-the-cuff remarks on the basis of what I’ve found in my data and what I have read:

  • Paying politicians is a double-edged sword:  increases in pay theoretically attract “better” types into politics, but they also attract “bad” types who are motivated solely by the money.  The idea is that these bad types have very poor options in the outside economy, but if political pay is high, then the bad type might as well “buy a lottery ticket” as it were and run for office.  So say Caselli and Morelli (2001). It then becomes an empirical question whether good types outnumber bad types among candidates, and if elections are somewhat random you may end up electing more bad types.
  • It’s actually very hard to assess the impact of improved pay on legislators because in most legislatures all members receive the same pay.  This makes it hard to disentangle changes effected by the improved pay from those effected by, e.g., simultaneous changes in the outside economy. (Certainly, we pay ministers & such more, but ministers get all sorts of other perks & influence that we can’t tell whether it’s the pay or the other perks that make minister different from backbenchers.)  There are some exceptions to this rule:  MEPs used to be paid what their respective national MPs were paid.  This meant that, e.g., Bulgarian MEPs  made less than German or French MEPs.  This rule was then changed so that the MEPs from lower-paying countries secured a significant raise.  Fisman et. al. (2012) looked at how MEPs responded to this change:  the results were in line with the Caselli and Morrelli predictions:  the educational level of MEPs elected in the effected countries declined and the number of candidate running increased in line with home country corruption, i.e., bad types were increasingly attracted to office.
  • My own analysis of the impact of pay increases on Canadian MPs suggests to me that the returns to increased pay in countries like Canada will be small.  The reason is that Canadian MPs’ pay levels already put them into something like the top-3% of income earners in Canada.  So any pay increases are likely to produce very marginal changes in behaviour.  Also, given the long right tail that characterizes the income distribution (especially in a place like London), you have to think about how much you are going to have to increase MPs salaries to effect visible behavioural change, x 2, x 3 of the current rate? And per my point above, how would you know if this pay increase worked?  It is precisely because these effects operate at the margin that I think that papers showing real & positive changes from paying politicians in, e.g., Brazilian municipalities (see Ferraz and Finan 2008), are of limited relevance to national legislators in advanced industrial democracies.
  • My intuition is that a strategy-proof contract would pay politicians sufficiently to make them indifferent between entering politics or continuing to operate in the outside economy. So we’d pay a MD or a corporate lawyer a lot more to be an MP than a school teacher.  Now, you’d probably have to offer a bit more compensation given that MPs travel extensively and must maintain a 2nd residence. For example, you’d have to pay me a LOT to trek from Vancouver to Ottawa in February.  Ottawa… Ugh.  But, then, somebody from say, Hull or Sheffield (apologies) might pay to be able to live and work in London. I’m just saying.
  • I think we ought to worry about just two things when it comes to paying politicians:  1) do we pay them sufficiently to ensure that Parliament is representative, especially in an economic sense.  Recall that MPs were not paid prior to about 1910, and the House was at that time the preserve of plutocrats – not good.  2) Do we pay politicians sufficiently to ensure probity in office?  This was the Economist’s argument – and there’s something to it.  However, as I pointed out above, simply raising pay may not achieve this end – you could just attract more bad types to Parliament.  It may be more effective to offer a lucrative pension subject to good behaviour & contingent on staying out of lobbying etc. after leaving politics.  I’d love to hear from some micro-economists about that idea.