Trouble at the top

I see in the news that CEOs are still getting pay rises bigger than their employees. CEOs pay has grown almost 10% in the past year, they say, compared with a 2.9% rise in average ordinary time earnings (and 1% consumer price inflation).

This is a trend that has been going on for quite a while, and it is much more of an issue in other places like the UK, where the government is requiring listed companies to have a shareholder vote on executive pay that will bind the company, and in the US, where recently the New York Times reported on 25 companies who paid their CEOs more than the companies paid in taxes. Cue dismay and disgust.

It does seem to be an issue all around the world, although these guys say it is mostly a US problem:

American chief executives received roughly four times what their Swedish counterparts in comparably sized companies did and 3.1 times that of a Japanese chief at a comparably sized company.

The average CEO at one of the 500 biggest US companies was paid $10.5 million a year in 2012, roughly split one third salary, one third other benefits, one third shares. Nice work if you can get it, although CEOs a bit longer in the tooth probably remember 2007 as the halcyon days – average US big-company CEO compensation peaked at $17 million each that year. And spare a special tear for poor Mr Dauman at Verizon who took home $84.5 million in 2010 but could only manage $43 million in 2011.

The ratio of CEO pay is obviously high enough to be a cause of significant political controversy, although CEO pay is not as high these days as it was earlier in the 2000s  relative to production worker pay on this data.

The golden goose

So why is this happening. Have we we got more companies that need leaders, driving up demand? Or perhaps fewer people coming out of CEO school looking for sandpits, reducing supply? Or maybe the value of what CEOs do has gone up over time somehow.

You can take your pick from a large number of hypotheses out there in the land of the commentariat. A selection of ideas follows.

  • Some, like these folk in the Wall Street Journal, argue CEOs generate their remuneration (which is often linked with share prices) from manipulating short-term results to drive up the value of their options, and don’t worry enough about long-term company health.
  • The Economist says that the increasing divergence between CEO pay and the pay of the unskilled worker is effectively due to globalisation. Referring to the UK, it says that CEOs with relevant skills are increasingly hard to find as companies diversify and internationalise (although presumably the HR department doesn’t lack for people willing to put their hat in the ring for $10 million a year). More soberingly, there is less to distinguish an unskilled worker in the UK from one in India or China these days, pushing down wages at the bottom end of the labour market as jobs move to where labour is cheapest.
  • Warren Buffett, billionaire investor, is on record as saying (see page 16 of his 2005 report to shareholders) “Too often, executive compensation in the U.S. is ridiculously out of line with performance”, and he particularly criticised large CEO exit payments:

Getting fired can produce a particularly bountiful payday for a CEO. Indeed, he can “earn” more in that single day, while cleaning out his desk, than an American worker earns in a lifetime of cleaning toilets.”

(Mr Buffett is not so indiscreet as to reveal how much time he has spent cleaning toilets in his lifetime).

  • The misalignment Mr Buffett talks about could be because compensation of executives in the US is more like a process of executives using their power to influence their own compensation and extract as much as they can rather than pay arrangements that are aimed at maximising shareholder value.
  • Or it might be because no Board wants to admit that they have a below average CEO by setting a below average pay package in this era when everyone knows what CEOs earn (although I think the world would be a better place if information on what people earned was more freely available).
  • Some say CEOs are mates with the people who set their pay and there is an unspoken rule that if they get approved higher salaries, then they will reciprocate when it comes to making the decisions for others. Or that those advising the relevant peeps on what the CEO should be paid are conflicted by other valuable work that they are doing for the CEO. Or that more generally, there just isn’t sufficient connection between company results and what the boss gets paid.
  • Paul Krugman, Nobel Prize winning economist, argues that the upsurge in executive pay was brought about by political and social factors, including an environment where the media and politicians are less likely to criticise inflated pay rates, weaker unions, and a sharp decline in marginal tax rates.
  • These academics argue that it is basically because the value of the companies CEOs run has gone up – similar to the Economist’s argument. So compensation for executives at the largest companies has risen because the market capitalisation of the largest companies has gone up. If you want to be paid more, work for a bigger company. Although reportedly business services, computers and banking are exceptions – they get paid more anyway. Leading some to say that “perhaps chief executives can add more value in more dynamic sectors”.

Or you might not really care why it happens, but just want it to stop.

Input to output ratio

Interestingly, precisely none of these people think high CEO pay has anything much to do with CEO performance itself. Which leads us on to the interesting question of whether CEOs actually matter at all.

And that in turn leads us on to this really interesting Atlantic article. I quote.

In their groundbreaking “Leadership and Organizational Performance: A Study of Large Corporations,” first published in 1972 in American Sociological Review, Stanley Lieberson and James O’Connor … asserted that the CEO’s influence was seldom decisive in a company’s performance…. “Industry effects,” such as the amount of available capital and the stability of the market, accounted for almost 30 percent of the variance in corporate profits. “Company effects,” such as the firm’s historical place in the corporate pecking order, explained about 23 percent. “CEO effects” explained just 14.5 percent. And even this impact should be viewed skeptically: it unavoidably bundles CEO actions that were genuinely smart and skillful with those that were merely lucky.

Which, if true, means that your task as the CEO is to get your company into an area that is growing quickly. Jeffrey Immelt, the CEO of General Electric (the third largest company in the world by some measures), is reported to have said “Not only could anyone have run GE in the 1990s, [a] dog could have run GE. A German shepherd could have run GE.” Previous CEO Jack Welch more or less agreed with this assessment.

Others are even more sceptical about the value of CEO. The article again:

James March, a management professor at Stanford, says that in any well-run company that’s conscientious about grooming its managers, candidates for the top job are so similar in their education, skills, and psychology as to be virtually interchangeable. All that matters is that someone be in charge. “Management may be extremely difficult and important even though managers are indistinguishable,” he writes. “It is hard to tell the difference between two different light bulbs also; but if you take all the light bulbs away, it is difficult to read in the dark.”

Eek. How much do we pay them again to keep the lights on?

Silver linings

The Atlantic eventually conclude with two interesting points:

  • One, it is important to ask not whether CEOs add any value, but when they add value, i.e., in what circumstances do you want a rockstar CEO and when would rather have someone less dramatic and ultimately less change-oriented.
  • Two, good CEOs can improve things a little bit. Bad CEOs can really stuff things up. I think anyone who has ever had a boss or been a boss knows this. It is much much easier to stop/ruin/doubt/undermine than it is to support/improve/really add something useful.

I note the irony in Atlantic feteing the then CEOs of Research In Motion, Mike Lazaridis and Jim Balsille, whose “newest BlackBerrys are flying off the shelves”. No more. This is Research In Motion’s share price  for the last five years (the article was written June 2009). Mr Balsille is particularly infamous for saying that he wasn’t worried about the iPhone, and didn’t think it would affect sales of Blackberrys. That said, they got the last laugh, to the tune of $12m between them when Research In Motion finally showed them the door.


Why can’t we all just get along

Incorporating a modest sort-of book review of The Company of Strangers, by Paul Seabright, and The Spirit Level, by Richard Wilkinson and Kate Pickett.

Hmmm. This has turned out to be longer than expected. A short summary is called for:

  • Trust between citizens in society is crucial to make life pleasurable.
  • Trust in strangers is also crucial for the functioning of the economy – buying things you want depends on trading with strangers to an extent unprecedented in history.
  • Trust is fragile and undermined by substantial differences between citizens, including income inequality. New Zealand has quite a high level of income inequality according to recent OECD figures.
  • If trust is so important, then we want to craft public policy in ways that increase societal trust. Income redistribution is certaintly going to part of this, but there must be more to the picture too. The challenge is that these are complex, long-term, cultural issues and cause and effect are highly murky.

There you go. On with the show.

Trust and be damned

Trust in your fellow citizens is really helpful in an enormous number of ways. If you doubt this, consider all the effort you go to every single day because you can not be sure that someone else won’t burgle your house while you are out, or steal your car, or take your bag or your phone from where you leave it on the table in that cafe you go to. And imagine how much easier life would be with fewer ID checks, security scans, credit checks, locks or keys, and less fraud, theft, or violent crime, let alone less fear. What would it be like if you could really trust your neighbours, let alone those with whom you have more distant social connections?

I treat “trust” as being a firm and justified belief in the reliability, truth and good intentions of a fellow citizen. My hypothesis is that societies where people trust each other more create fewer conflicts between citizens, minimise protective behaviours (like locking doors) that are only necessary because of a lack of trust, and resolve conflicts that are created much more easily and positively. There is less mis-understanding, more helping out your fellow humans, and a higher level of what we might call “societal empathy”, which I think of as the ability to understand, share, and act as if you care about the feelings of someone else with whom you share space in society. As a result of this enhanced low-level (person to person) co-operation, there is less need for higher level (government to person) intervention, and society just becomes a nicer place to be.

I have read a couple of interesting books recently that touch on this issue of societal trust. Both authors argued that trust in society can not be taken for granted, and is undermined by substantial differences between citizens. This makes sense to me. You can’t trust people that you don’t empathise with, and empathy requires having some interaction with other people or some shared experience that makes you feel like you have something in common, and makes you willing to do something to help another person.

Friends with benefits

Paul Seabright’s “The Company of Strangers” tells the remarkable (but mostly unremarked) tale of how much your life depends on getting along with strangers. Mr Seabright thinks that we have reached the point where even the most basic activity relies on a web of international co-operation that functions without anyone being in charge overall. Just think about how you get the food you eat (probably in a supermarket – grown, processed and brought there from all over the world by people you will never meet), the clothes you are wearing (ditto – and even more complex than fresh food because it has a longer shelf-life), and the device you are reading this blog post on (likely built in Asia to a foreign design with components from all over the world, and charged using electricity brought to you through a complex process run entirely by strangers who have no idea who you are, or what you are using their product for).

Given how well this system works most of the time, one could be forgiven for thinking that it is the natural order of things. Mr Seabright says otherwise, reviewing the evolutionary history that has enabled humans, alone amongst all species, to come to trust strangers enough to co-operate on everyday activities to such a extraordinary extent. For most of human history strangers represented danger; they were to be avoided or treated with grave suspicion. Trading between unrelated people was a difficult and unpredictable business, and meetings between strangers could just as easily end in violence as in a mutually satisfactory exchange. But, once we really got the hang of it – which has only been in the last few hundred years – this trust and the trading that has resulted has enabled remarkable improvement in people’s lives all over the world, (as well, of course, as leading to a bunch of other developments of more questionable social value).

The trust system relies on a network of “institutions” – in the economist’s sense of sets of rules, both formal and informal that govern social behaviour – that mean that humans generally treat strangers as honorary friends. We are prepared to deal with strangers in accordance with the rules in the expectation that they too will follow the rules, and the rules or institutions that have been developed are robust, i.e., reliable enough to be taken for granted most of the time, and self-reinforcing to a large extent, i.e., in a variety of ways citizens punish those who betray others’ trust.

The economic system also works because no one really has to think too deeply about the systemic consequences of the actions they take in their little piece of the world. People proceed simply to make decisions that seem to them the right ones within a framework of rules set by the society in which they dwell, and the insititutions as a whole emerge from that behaviour – what we might call “short-sighted co-operation”. To use the example of a trading situation, I can sell you whatever it is that I make without worrying too much about the use you are going to put it to. As the extreme cases, Mr Seabright talks about firms that make cluster munitions, or people who work in nuclear weapons laboratories. In both cases, employees simply do not spend a lot of time considering the negative consequences of their jobs, but like most others, they focus on doing the job they have as best they can. These blinkers are not required in order to make the economic system work, but they do make it a lot more effective because they enable a lot more trading to happen. Consider the time and difficulty of buying a television on hire purchase (where you need to establish your creditworthiness) compared with paying cash for the same item (where you are relying on the credibility of the bank that issues your currency). If one had to be reassured of the good intentions and positive impact of all buying and selling, our market system would work vastly less effectively.

Mr Seabright discusses the limitations of the trust system. It can break down because trusted people turn out to be untrustworthy, or because the short-sighted co-operation leads to society doing things that are unsustainable for society as a whole. He talks about a wide variety of different forms of societal co-operation – cities, environmental issues, the market system itself, firms and families, the growth of knowledge, and also about social and economic exclusion in the form of unemployment, poverty and illness – and discusses why they have emerged, the constraints and difficulties they present, and the prospects for the future.

The book talks in some detail about the system of money and about banking, and discusses in particular lessons that can be drawn from financial crises. There is a particular application to the events of 2007 when, as the real estate market turned sour in the United States, banks quickly became extremely unwilling to lend money to each other, uncertain as to the others’ exposure to what were sure to be substantial property losses. The speed with which the disaster evolved and the extent of the damage it did – leading to both wide-ranging (and expensive) intervention by the US Federal government in the financial markets, and enormous financial losses for shareholders in affected banks – just emphasises how amazing it is that we put so much faith in the financial system as a normal part of our everyday lives. If my experience is typical, we do not really think at all about the risks we are taking in depositing our money in a bank, or borrowing someone else’s money via a bank to buy something that we want without having to save what it costs, like a house.

In the last section of the book, Mr Seabright discusses the development of collective action that societies take to respond to some of the perceived short-comings of the market system, both within their own borders, and in co-operation with other nations. And in the last twenty pages or so, he discusses how fragile this reliance on strangers is, particularly in terms of relations between states and what the prognosis is for the future.

Mr Seabright’s book is very good on economics: indeed, the book is in part a clever re-telling of basic micro-economic theory – when markets work well, when they don’t, what can be done about it in either case. His book is less focused on issues apart from market-based exchange, like the importance of trust and reliance on strangers for community life and the enjoyability of life in cities.

Nevertheless, I think his basic point is extensible beyond economics: modern life requires thorough-going trust of strangers, so modern life is going to get more and more difficult if people trust each other less. If this is right, then public policies that increase societal trust could be useful and indeed increasingly necessary.

Esprit de corps

“The Spirit Level”, by Richard Wilkinson and Kate Pickett focuses on income inequality in societies, and links higher levels of income inequality with a very wide range of societal ills. Their basic point, substantiated with a great deal of data, is that societies with less equal income distributions do worse in a very great number of ways. They have worse mental and physical health, higher rates of obesity, lower life expectancy, worse educational performance, and higher rates of teenage pregnancy. They are also more violent, more crime ridden and suffer from less social mobility. This is shown both across countries and, for many of the same measures, across states in the United States with varying levels of income inequality.

Note that the negative effects referred to are not because countries or states are poorer than their comparators. They relate just to the level of income inequality, regardless of the total income level. Higher incomes are associated with higher life expectancy, to take just one example, but at any level of income, a less equal income distribution will mean lower life expectancy for the population. In this way the social ills tracked are shown to be related directly with income inequality.

Also note that the negative effects of income inequality are not just confined to the poor. It is true that, on many of the measures, poorer people in a society do worse, with lower life expectancy and higher rates of imprisonment, for example, that those with higher incomes. But, from the data, even those on high incomes are better off in a society with a more equal distribution of income. Greater equality helps the poor most, but helps all levels of society to some extent (except perhaps, say the authors, those who are very rich).

The scale of these differences between societies is enormous. As the authors say (p181 in my hard-back version), across whole populations, rates of mental illness are five times higher in the most unequal compared to the least unequal societies. People are five times more likely to be imprisoned, six times as likely to be clinically obese, and much more likely to be murdered. For all of a large number of social ills, living in a society with more equal income distribution means you will be socially far better off.

Put another way (p261), if the United States were to reduce its income inequality to about the average of the four most equal of the rich countries (Japan, Norway, Sweden and Finland on The Spirit Level data), rates of mental illness and obesity might be cut by two-thirds, teenage birth rates could be more than halved, prison populations could fall 75%, and people could live longer while working around two months less a year. At the same time, the proportion of the population feeling they could trust others might rise by 75%, with obvious positive consequences for the quality of community life.

Interestingly the book also refers directly to the impact of higher income inequality on trust within societies. There is long-term evidence quoted in the book that greater levels of income inequality reduce the extent to which people feel they can trust their fellow citizens and the extent to which they are willing to look out for them. In the terms of my hypothesis, greater income inequality makes citizens feel they have less in common with their fellow travellers, and this makes them less inclined to do anything to help others out.

How much equality

Usefully , there has been some news media recently about New Zealand’s relative level of income inequality. An OECD study has found that New Zealand and Sweden (!) have bolted up the international rankings for income inequality in the last 25 years. Most of the change apparently happened in the period from 1988 to the mid-1990s, with the nation above the OECD average level of inequality since then despite incomes becoming slightly less unequal since 1995.

The summary of the OECD study itself says several interesting things:

  • Income inequality has slightly declined since 1995 after growing sharply for the ten years before that. There has been income growth for all, but the middle class has been doing particularly well, reducing inequality in the income distribution.
  • Poverty has grown throughout the last 25 years to levels similar to the rest of the OECD (about 11% of people are now living on less than half of the median income, which is the definition the OECD uses for poverty).
  • In terms of distribution, the proportion of children living in poverty has grown to 15% – at the higher end of OECD figures – and poverty among younger adults has also grown. Poverty amongst the elderly, on the other hand, is amongst the lowest in the OECD, at 2%.
  • Poverty is highly correlated with unemployment. Almost half of people living in households where no one works are poor, compared with 4% of households in poverty if there are two or more people working.*

If societies with more equal income distributions are better places to live, then New Zealand became a less nice place to live through until the mid-1990s, and since then it has slightly improved. I guestimate that the sharp decline in equality after 1985 was caused in large part by sharply rising unemployment, which peaked in the late 1990s and then fell away to very low levels in the 2000s.


In terms of politics, the extent to which people think increasing inequality is a good thing or a bad thing varies quite a lot between even right-thinking individuals. And many people don’t like the idea of not liking inequality because it implies that they have to do something about it – especially by taking money from people who have more and giving it to people who have less.

I think some inequality is useful and important, but too much is likely to be bad, in the sense of doing more harm to society than good.

The economically useful part is when differences in income between different people reflect differences in the value of different jobs. These differences help guide people in making choices about what work they want to devote themselves to. Surgeons command higher salaries than baristas in large part because society is willing to pay more for quality surgery than they are for quality coffee. The differences in salaries between these two groups will help ensure that roughly the right number of people work as surgeons relative to the number that work as baristas. Income differences can be pro-social, i.e., they work for the good of society, in these circumstances.

Human beings, however, seem to have very strong views about fairness that limit appropriate distributions of income between different people. Some concept of fair play is built firmly into society, and so it seems normal for people to express concern and want to do something politically about income differences that are not justified or justifiable (my hypothesis is that this justification is made or should be made on the basis of the usefulness of a particular income distribution, from a society-wide point of view). We seem willing as a society to tolerate a level of income inequality, but once it gets out of hand, there is strong societal discomfort, condemnation, and public policy action to put in place a more even distribution. A lot of the debate that we see about economic policy seems to operate in these terms. Note that this is in a country where people generally do not get paid much by rich world standards, where the salary rates for the best paid do not reach the stratospheric levels of other nations, and where the systems for income redistribution are already very extensive.

What is to be done

If we are going to make public policy with a closer eye on how it might affect the level of trust between citizens, it is clearly going to require a much better understanding of what makes people care about others. As Rudyard Kipling wrote in “We and They”, quoted in The Spirit Level “All the people like us are We, and every one else is They”. If it is a good thing for society that everyone look out for each other more, how should it set things up so that people empathise with their fellow citizens?

The making of “they” can be incredibly powerful and I you see it all around me all the time. A person who receives public assistance is either one of us fallen on hard times, as the Dom Post had it the other day, or they are one of them – a work-shy dole-bludger in need of harsh treatment for his/her moral failings. Mr Seabright recounts an ingenious experiment from Benin in 2001 (p294) that demonstrated that voters would prefer policies that enriched them personally at the expense of neighbouring citizens over policies that would be of more general benefit to the nation. The tendency for politics especially to tribally appeal to supporters and define all others as different and dangerous is a constant threat even in liberal democracies like the one in which I am fortunate enough to dwell.

So the question is how do we encourage people in society to look out for each other, and not just those who are the same as us (although that would be a good start), but others who are different but not less worthy of our respect and, yes, assistance, because of that.

On my read, both books are less good on solutions and the policy agenda than they are on establishing their basic points. I am sure this reflects the long-term, uncertain and difficult nature of these problems – which therefore means they are more amenable to political solutions rather than careful policy analysis that weighs up the consequences of alternative approaches.

But from the books, some practical steps to build a more trusting society might include:

  • Redistribute earned income, i.e., try to ensure that people get a more equal recompense from working. The most obvious way to do this is with policies that reduce unemployment, since unemployment is closely tied with poverty, and also policies that raise wages, i.e., encourage the nation to focus on producing things that the world is willing to pay a lot for, and ensure the education system makes the maximum contribution to educating people for high-wage work.As part of this redistribution, The Spirit Level also suggests constraining the forces that generate inequality, in particular the extremely high levels of CEO pay, which perhaps is less an issue in New Zealand. The British government is talking about introducing a simple mechanism for shareholders to vote on CEO pay, intending that this will give more direct control by company owners.

    The Spirit Level also suggests employee ownership as a solution (rather than such extensive ownership by investors with only a financial stake in the company. They argue that more employee ownership would encourage a more long-term, less rapacious and more fulfilling (for employees) style of capitalism if it were combined with more participatory management methods.

  • Redistribute disposable income – i.e., a more progressive welfare system. It isn’t clear to me where the limits of redistribution lie, but it does seem to me that New Zealand is quite progressive already, with 70% of the tax being paid by 25% of the taxpayers. Other tax changes, like introducing a higher rate of income tax for those with very high incomes, seem to be mostly of symbolic effect – they don’t raise enough money to change the distribution of taxation very much at all.
  • Provision of public services with public money, especially health, education and welfare services, can help even up society by effectively insuring citizens against the financial burden of education, ill-health or economic or social crisis. The quality of these services also matter: Mr Seabright discusses the importance of a “secular, multi-ethnic and liberal” education to ensure that citizens learn “how to live peacefully and profitably with people whose community and religion are not one’s own”.
  • Socialise these ideas – The education system is far from being the only place where people learn how to behave. This point is really just a sustained effort to convince people that the world can easily be a different and more pleasurable place to live, and that improving things is partly down to them and how they live their lives every day. So get to know your neighbours, start a after-school programme, run a block party, join a community group. Get amongst it citizens, the health of your nation is at stake.

* The OECD also mentions that New Zealand does a good job of targetting its welfare at the poorest – one third of total cash benefits go to the poorest 20% of the population – only Denmark and Australia do better. Which is probably just as well, given how little money our nation has to distribute by comparison with the rest of the (much richer) rich world. (It amazes me that managing to get only one third of cash benefits targetted at the poorest 20% is considered a good performance – however, that is for another day).

** These specifics come from The Spirit Level, where the game is discussed from page 199.

Taxing the rich

There was lots of talk about taxing higher income earners more in the election campaign. In particular:*

  • Labour said it would introduce a new tax rate of 39 cents in the dollar for income over 150k (currently 33 cents), and reduce to zero the tax rate on income below 5k (currently 10.5%).
  • The Greens proposed higher income tax rates at all income levels (including a 39 cent rate on income over 80k), except for those earning less than 10k, who would pay no income tax.

To some extent both proposals seem to be based on an argument that taxing those on higher incomes is fairer, and that at present the lowest income earners are too highly taxed. The Green party proposes to raise income tax for everyone except those earning less than 10k – so the justification for its proposal would presumably also be that higher taxes in general are useful and important to support other policy priorities.

It is worth noting though, that those on higher incomes already pay a lot more income tax than others. This is not just because they earn more, but also of course because the income tax rates get higher as incomes grow. The current tax rates start at 10.5% for incomes up to 14k, and go up to 33% for incomes over 70k. So anyone who earns more than 70k pays three times as much income tax for every extra dollar s/he earns than those who earn less than 14k.

It is interesting to see how these rates shake out in practice, i.e., applied to  incomes for real people. The Treasury releases the helpful “Facts for Taxpayers” with every Budget in May. From the table for the 2011 Budget, I have made this slightly simplified version of who earns how much and how much tax they pay.

Annual taxable income (Budget 2011)**

Number of people Amount of tax
(000) % (m) %
<10k 677 20 221 1
<50k 2519 75 6903 30
>50k 823 25 15907 70
>100k 164 5 6958 31

Some facts are immediately obvious:

  • Incomes are low – 75% of those who pay income tax earned less than 50k in 2011 – this is 2.5 million people out of total of 3.3 million taxpayers.
  • The tax system is quite progressive – i.e., the more you earn, the higher the proportion of your salary that you contribute. The 25% of people on more than 50k paid 70% of the tax in 2011. The 20% of taxpayers on less than 10k (which includes 244k people who have a zero income) pay 1% of the tax.

I had a look at the same numbers since 2002. These tables show the proportion of people in each band over time, and the proportion of tax they pay. If I had the genius of Keith Ng, I would do a data visualisation but sadly I do not.

Proportion of people in each salary band by year (%)

2011 2010 2009 2008 2007 2006 2005 2004 2003 2002
Salary band
<10k 20 20 21 21 21 22 22 23 25 26
<50k 75 76 79 78 80 83 84 87 86 87
>50k 25 24 21 22 20 17 16 15 13 14
>100k 5 5 3 5 3 3 3 3 2 3

Proportion of income tax paid by people in each salary band by year (%)

2011 2010 2009 2008 2007 2006 2005 2004 2003 2002
Salary band
<10k 1 1 1 1 1 1 1 2 2 2
<50k 30 32 36 34 37 39 43 45 45 47
>50k 70 68 64 64 62 60 57 55 54 53
>100k 31 29 26 29 27 27 25 24 24 24

We can see some interesting trends:

  • Incomes are growing over time – The number of taxpayers earning more than 50k grew 7% a year from 2002-2011. As a proportion, taxpayers in the more than 50k bracket have grown from 14% in 2002 to 25% in 2011.
  • Tax incidence is growing more progressive – Those earning less than 10k (including those with zero income) paid 1% of tax in 2011 – down from 2% in 2002. The proportion of tax paid by those on more than 50k is up from 53% in 2002 to 70% in 2011. Those on more than 100k pay 31% of tax in 2011, up from 24% in 2002.
  • National’s tax cuts in 2010 do not seem to have made the tax system more regressive. On the contrary, the porportion of tax paid by higher income earners has gone up from 64% in 2008 to 70% in 2011.

Presumably a lot of this extra progression (i.e., the higher and higher proportion of tax paid by those earning more than 50k) reflects growth in incomes over time. The number of people earning more than 50k has grown 8% a year since 2002, while the number of people earning less than 50k (including those who earn nothing) has fallen a gentle 1%.

We can try to compensate for this growing incomes effect by calculating the amount of tax paid per person across income bands. This is just the total income tax for each band, divided by the number of people in that band. The results are shown below.

Average amount of income tax paid per person in each salary band by year ($)

2011 2010 2009 2008 2007 2006 2005 2004 2003 2002
Salary band
<10k 326 348 373 364 388 468 462 474 553 547
<50k 2740 2937 3246 3492 3753 3502 3491 3454 3393 3393
>50k 19328 20635 20917 24480 24884 25807 25679 25281 25030 24997
>100k 42427 49106 49975 56420 57407 63042 62294 62179 61915 58870

You can see that the average amount of tax paid has fallen across all income bands since National came into power in 2008. For those on less than 10k (which includes those with zero income), this is a continuation of a long trend – tax has fallen from 547 in 2002 to 326 this year.

You can also see the impact of National’s income tax cuts:

  • Those earning less than 50k (including those with zero income) have seen their tax fall from an average of $3492 in 2008 to $2740 this year.
  • The average income tax paid by those earning more than 50k has fallen from 25k in 2008 to 19k in 2011.

From this you can argue that most of the tax cuts, in an absolute sense, went to those who earn the most. So the total tax paid by those earning more than 100k has fallen 14k (from 56k to 42k) since 2008, compared with the total tax paid by those earning less than 50k falling by $500 (from 3.2k to 2.7k). But this is a direct result of the progressivity of the tax system – i.e., tax cuts mostly benefit higher income earners because higher income earners pay most of the tax.

As we saw above, in practice the proportion of tax paid by higher income earners has grown over time rather than fallen. And the tax cuts have been reasonably even spread – the average amount of tax paid by those earning more than 50k fell 21% between 2008 and 2011, compared with 22% for those earning less than 50k. Those earning more than 100k did a bit better – their average income tax bill fell 25% – reflecting the impact of removing the top tax rate.

You can download the data I compiled from the Treasury’s Key Facts for Taxpayers. I can send you the documents from 2000-2006 as well if you like (the 2007-2011 versions are on the Treasury website and the Treasury helpfully sent me the 2000-2006 versions). Drop me a line on twitter or use the contact form.


* There is a lot more to both parties’ tax policies than this – the focus here is just on income tax.

** The Treasury says that these numbers include tax on NZ Super and major benefits, and exclude ACC levies, Working for Families tax credits, and anyone under 15. Data is projected for the year ended March based on the Household Economic Survey from Statistics New Zealand. The Ministry of Social Development says that nearly all households earning under $70,000 a year are eligible for Working for Families. So I would expect that higher income earners would be paying an even higher proportion of tax
if these tax credits were included in the numbers.