How should we remunerate politicians?

Many of us spend a lot of time bashing politicians – sometimes fairly, sometimes not – but it cannot be denied that the job which they do in most democratic societies is a thankless one. They have to subject themselves to electoral approval every four or five years and they get a lot more brickbats than bouquets – when things go wrong, it is politicians who stand in the front line of public criticism. You have to be idealistic enough to believe you can change things; extrovert (some might say narcissistic) enough put yourself forward to the electorate on a regular basis and sufficiently thick-skinned to cope with the criticism. It puzzles me why anyone would ever want to do that – in most developed economies, at least, the rewards do not stack up to the degree of opprobrium heaped upon politicians.

Aside from the prime minister of Singapore, who earns an annual salary of USD 2.2 million, public sector chief executives are paid significantly less than chief execs in the private sector (see chart 1). There are various metrics against which we could measure a country’s performance in order to assess whether the chief executive is worth their pay. One way would simply be the size of the economy – the bigger they are, the more they can afford to pay. But that rather falls at the first hurdle given the size of the Singaporean economy. Indeed, in our sample of 60 countries, the correlation between the ranking of overall economy size (GDP) and the salary of head of government is rather low, at 0.375. Another way to think about it is that the bigger the population, the higher the salary should be given that they are responsible for many more people. But this falls pretty flat with the rankings showing zero correlation – hardly surprising given that President Xi of China earns only USD 22,000 per year and Indian Prime Minister Modi’s official salary is just USD 30,000.

We get more success, however, if we look at incomes per capita with the rankings showing a correlation of 0.55 (rising to 0.57 if we look at PPP adjusted incomes). Societies appear prepared to reward their politicians on the basis of their ability to deliver economic prosperity. Naturally, there are some outliers. The President of the Comoros earns a whopping $408,000 per year (541 times the average income) despite it being one of the poorest countries in the world. Contrast this with Uruguay where the presidential salary is only 80% of the average income. Indeed, the degree of disparity is highest in countries not known for the quality of the democratic process (chart 2). But on the whole there is a decent degree of correlation: For example, Donald Trump is paid the fifth highest salary (USD 400,000) whilst the US ranks sixth in terms of GDP per capita; Theresa May’s salary of USD 215,000 puts her 13^th on the list compared to a GDP per capita ranking of 16 whilst French prime minister Édouard Philippe’s salary of USD 200,000 is placed 16 against a GDP per capita ranking of 17. Perhaps societies are indeed prepared to reward their leaders in proportion to how well they manage the economy.

Even the position in Singapore, where the apparently outlandish salary of PM Lee Hsien Loong is off the charts on an international comparison, does correlate with the fact that standards of living are amongst the world’s highest. Singapore is also safe, clean and ranks second on both the WEF’s global index and the World Bank’s Ease of Doing Business index. At a time when the economic threat posed by China is increasing, that is no mean feat.

But compare this with corporate salaries where the median total compensation for S&P500 CEOs in 2016 was USD 11.5 million whilst the average FTSE100 CEO received remuneration of around USD 6 million. Thus, given the dissatisfaction with politicians around the world, would we improve their quality if we paid them more? The case for paying more is that we attract a higher quality of candidate. But there is also a principal-agent problem: Politicians generally set their own salaries, and aside from the odd burst of outrage, in many countries there is little to stop them from raising their own salaries in excess of inflation or indeed in excess of their deliverable outcome in a way which is detrimental to society.

The empirical evidence is mixed. A 2009 paper based on Brazilian data suggested “that higher wages increases political competition and improves the quality of legislators.” Similarly, a study based on Mexican data  indicated that the quality of civil servants also increases when higher wages are offered. Against that, a study based on the European Parliament found that a salary hike raised the proportion of politicians with lower education levels, perhaps because  it “gave lower-quality MEPs a greater incentive to get into office and, once there, to stay put.”

As Stephen J. Dubner of Freakonomics fame put it, “I am not willing to argue that paying … government officials more would necessarily improve our political system. But, just as it seems a bad idea to pay a schoolteacher less than a commensurately talented person can make in other fields, it is probably a bad idea to expect that enough good politicians and civil servants will fill those jobs even though they can make a lot more money doing something else.” There are no easy answers to the question of how to reward politicians. But we generally know a good one when we see them.