Feed the Leeches More Blood - Get Better Leeches
I am doing a personal research project on Singapore and I noticed an interesting bit of information: in 2007, the Singapore government came under criticism when it increased the salary of its Prime Minister to about $2 million. That is certainly very different from the American philosophy on politician compensation, which holds that it should be as small as possible. Our President is paid a salary of $400,000 and our congressmen all make less than $200,000.
On a possibly related note, international organizations regularly rank Singapore as one of the least corrupt countries in the world, while the US does a little worse (however, it is still pretty good). I wonder if paying our politicians so little contributes to corruption. After all, there must be some reason why presidential candidates were willing to spend $1 billion this year to get a position that pays only $400,000, or why Rod Blagojevich was offered $500,000 for Obama's senate seat which carries a salary of less than half that sum.
The optimistic take is that the warm fuzzy feeling of public service (or more realistically, prestige) is so powerful that people are willing to sacrifice to get it. The pessimist in me thinks that they do it for non-direct benefits of the job, including future lucrative positions with lobbyist firms and industry groups that they help out while in office.
Perhaps paying politicians more would reduce the incentive for them to deal under the table. When I see the salary of the Singapore Prime Minister, I am reminded of the salary of a CEO. When I see the pay package of our politicians, I am reminded of a college athlete: they are contractually bound to receive a low compensation, but your star center didn't buy those rims with his momma's money.
Our low compensation structure reduces the appearance of corruption (and Americans have something against people that make high salaries), but it increases the probability of actual corruption.
More formally, let's define an action by a government official to be objective when his purpose is to act for the good of society (whether or not he is correct). Then define an action by a government official to be corrupt when his purpose is to use his position for private gain. The principal of good governance would hold that we should try to increase the ratio of objective actions to corrupt actions by government officials.
In this framework, an official should be well-compensated to the point so the utility of corrupt act is small. This is especially true if the cost to society of a corrupt act is large compared to the personal gain of the official. It would be cheaper to just pay him what he would have made by dealing under the table, straight from the public treasury.
Perhaps high-level officials should even be guaranteed public jobs for life to reduce the incentive to lobby for corrupt acts from other officials after they leave office. An ex-official that siphons off $billions for a lobbying firm every year is surely more expensive than one that is put up in an office somewhere in the Washington catacombs, sipping a coffee and bossing around an intern. Perhaps this is the true purpose of Presidential Libraries.
Best of all would be to eliminate the difference between public benefit and private benefit for each official by paying them according to their performance. Many industries have this already, like football quarterbacks. They are paid extra if they throw a lot of touchdowns or if their team makes the playoffs. Unfortunately, I don't know how you would design an incentive scheme for, say, a legislator. Perhaps you could pay them based on net migration to the jurisdiction they legislate for.
It strikes me that in thinking up ways to make government better, I am actually just mimicking the thought process that the compensation committee in a private government-firm would go through in some sort of market anarchy.