Tip of the iceberg

What underpins top pay?

University of Auckland Business Review, Spring 2016
Graphics: Ester Tongs
LENGTH: 9 min (2283 words)

> The search to discover what drives CEO remuneration has a long way to go.

The lives of the men and women who run our largest organisations  – the Chief Executive Officers, or CEOs – are both public and mysterious.  We know who they are, though few observers could name their immediate subordinates – the CFOs and COOs – in even the largest companies. In the case of publicly-listed and state-owned companies, we also know how much CEOs  are paid, and we are periodically and very publicly reminded how large their remuneration is, and how much it has gone up – as it always seems to do – compared with that of other wage and salary earners.

What is still mysterious, however, is why CEOs are paid what they are; why pay differs so much across firms and countries; and why, in particular, top pay across the board has about doubled in New Zealand over the past two decades.

To put it bluntly: are CEOs worth what they get paid? If the answer to that is yes, were they being exploited when paid so much less twenty years ago? If the answer is no, then how has the difficulty of the job or the value of the output increased so as to justify the higher pay?

One reasonably concise proposition about the actual job of a top executive is advanced by researchers Peter Demerjian, Baruch Lev, and Sarah McVay in a 2012 Management Science article. In it, they write that they:

expect more able managers to better understand technology and industry trends, reliably predict product demand, invest in higher value projects, and manage their employees more efficiently than less able managers. In short, we expect more able managers to generate higher revenue for a given level of resources or, conversely, to minimize the resources used for a given level of revenue (i.e. to maximize the efficiency of the resources used).

This is certainly language any economist can understand, but can economic analysis reliably identify and measure the efficiency of managers' use of resources, and empirically link this to their (generally increasing) levels of remuneration? The present article investigates the topic by drawing on research that uses novel – indeed, unique – data to identify the deep structure of managers' pay in New Zealand.

Since the 1993 revision of the Companies Act, all firms listed on the New Zealand stock exchange that are substantially locally owned have been required to include in their annual reports information on all employees with remuneration packages worth more than $100,000 per year.

This data, presented in the form of salary bands, enables us to infer the size and shape of the managerial hierarchy within each firm. It can be added to more conventionally available information on the structure and performance of firms, such as sales, assets, number of employees, profitability, and growth, to give new insights into what CEOs are paid to do, and what might justify – or, at least, explain – why their pay has tended to increase over time.

In 2014, CEOs in the New Zealand listed companies sector received an average compensation of $840,000 – a figure that has increased since 1995 by 85 per cent in constant dollars, even when corrected for company size. Over the same period, real weekly wage and salary earnings have increased by just 13.5 per cent. As a result, CEO pay is now on average 12 times higher than that of the average shop-floor worker.

The “explosion” (as some have termed it) in top pay has been documented across the English-speaking corporate world, and indeed the process probably got underway well before 1995. In New Zealand, for example, the income share of the top 1 per cent of taxpayers almost doubled between 1985 and 1995.

So, how has this happened, and is it a problem for the firms themselves – that is, a governance or capital market issue – or indeed for the nation at large, through its effects on the overall income distribution and economic prosperity?

Here is where the mystery becomes evident. No one seems to know why CEOs are paid so much – if indeed it is a lot –  because we do not know exactly what their jobs entail. That is, we do not know what tasks and responsibilities CEOs are being paid to shoulder, so we can hardly assess how effective they are, or how much that activity should be worth to the companies that employ them.

This may seem a surprising claim, because the pay and performance of CEOs – and sometimes of their most senior executives – has received considerable attention from researchers. These researchers have adopted one of two analytical perspectives: changes in what could be termed “horizontal”  bargaining power of (especially English-speaking) CEOs with globalisation and benchmarking, and changes in the “upward-vertical” relationship between CEOs and their boards, as reflected in the structuring of remuneration packages

These are, of course, important issues, but they leave unexplored another possibility. What if CEO pay actually has not risen much, if at all, when organisational structure is taken into account. The reasoning might go as follows: along with providing overall leadership and signing off on major resource allocation decisions, CEOs have the job of “managing” their companies. In practice, this has come to mean that the CEO directly supervises a small number of senior executives who report directly to him or her, and they in turn supervise their own cadres of less senior managers, and so on down to the supervisors on the shop or office floor. That is, most or all listed companies are organised as hierarchical bureaucracies.

Researchers agree that employees at higher levels in those bureaucracies should, and do, get paid more – even if there is disagreement about just what they are being paid more to do. That is, managers at each level get paid more than the people they supervise, and less than those to whom they report.  So, if – for reasons yet to be explored – there has been a trend to larger managerial bureaucracies (for a given revenue size of firm), then this would naturally be reflected in increased pay for the person at the top of the pyramid. On the other hand, if the trend is to flatter bureaucracies, or wider spans of control, then it is conceivable that the greater skill required to manage such structures might equally justify higher pay for the CEO.

What the data looks like

Information was collected from the annual reports of locally-listed companies using the financial statements collected on the NZX database (, often supplemented by the text of the reports – for example, to glean figures on employee numbers. For each company, data were obtained for the latest available financial year (usually 2014, when these data were gathered), and the earliest year in which the newly required remuneration numbers were offered (usually 1997).

The highest-paid CEO (at telecoms company Spark) received a remuneration package in 2014 worth $3.8million. The lowest paid, at start-up Windflow Technology in 2005, earned just $110,000. Any idea that there is a rigid relation between company size and CEO must be qualified by the fact that the total revenue of Spark in 2014, at just under $3.7 billion, was less than half that of the largest listed company by revenue, Fletcher Building ($8.4 billion), even though, at $1.9 million, Fletcher Building’s CEO was paid only one half of the Spark boss. Fletcher Building also had the most staff over the threshold of $150K annual remuneration, and the largest salary bill for those staff – 1078 people collectively earning almost $260 million.

Just over 5 per cent of our listed companies are in the “FIRE” (Finance, Insurance, Real Estate) business – a proportion that is probably lower than in the United States, Britain, and elsewhere. One quarter of the companies earned a better than 15 per cent rate of return (profit/assets), but nearly one in five were in loss-making territory – including some recent start-ups.

Although there were 110 listed companies with two years of data, most of these entered the database too recently, or departed too early to yield valid before/after comparisons. Table 1 shows averages for the 41 companies for which we have two years of data, and for which the earliest data year preceded 2001 and the latest was after 2011, with a sample average gap of sixteen years. All monetary data are converted to 2014 values.

The first data column is the average before the early year; the second shows the latest year, and the third column divides the second by the first, to tell us the average growth over time of each variable.

Here, "Size" is the larger of revenue from sales and total cost of sales. For these companies size increased on average by more than the growth in assets and total employment, which in turn grew more than real value added (sales minus cost of materials and purchased services), implying an overall drop in productivity. The average wages or salaries of both non-managerial and managerial employees barely increased, in sharp contrast to the more than doubling of CEO remuneration. However, the largest growth was in the number of managers, which supports the hypothesis that some CEO pay growth is due to an increase in the number of managerial subordinates reporting to them. Over the sixteen year average span between the first and second observations for these 41 companies, the ratio of managers to firm size increased, on average, by 122 per cent  – a striking change in organisational structure.

Modelling top pay

To dig deeper into the connections between pay, size, managers, and so on, econometric models were run on the full sample of 110 companies for which two years of data were available. These exercises can produce estimates of the magnitude and causation of linkages between variables, and they have striking implications:

  • Even controlling for changes and differences in firm size, there is an unexplained secular upward trend in CEO pay in New Zealand listed companies of about 3 per cent per year.
  • Size itself is important, as expected. On average, a company that is twice as large as another in terms of sales will pay its CEO 30 per cent more. This matches the findings in overseas studies.
  • About one third of both size effect and time trend can be accounted for by expansion in the numbers of managers – that is, employees paid more than $150,000 per year, most of whom are likely to be managers or to have substantial managerial responsibilities. However, it is not known why, or to what good, the managerial ranks have been inflated.
  • The number of employees in a firm earning less than $150,000 per year has no discernible effect on remuneration of that firm’s CEO. Whatever CEOs are paid to do apparently does not included taking responsibility for their blue- and pink-collar workforces.
  • Controlling for other factors, CEOs in the “FIRE”  (Finance, Insurance and Real Estate) sector are about 30 per cent better rewarded than their peers in other sectors.
  • A company’s profitability correlates with top pay only at the extremes: compared with everyone else, CEOs in the highest rate-of-return companies, and CEOs in companies making losses are the best paid, other things being equal.

Implications and further research

The striking secular upwards trend in the pay of people running New Zealand's publicly listed companies lacks any obvious empirical justification in terms of general improvements in marketing, productivity, and profitability.

Indeed, if anything, the mystery is deepened. We can attribute – or at least link – about one third of CEO pay growth to a fattening of the hierarchy of managers reporting to the CEO, directly or indirectly. But what is the cause of this apparent sharp decline in managerial productivity?

Future research will focus on these issues. Financial and remuneration data have been collected from annual reports for every year, and personal characteristics and work histories compiled for each CEO. Industry-level data will enable the benchmarking of each firm’s sales, productivity, and profitability performance to its peers, and the investigating of links (or the absence of links) between performance and top pay.

Possible ancillary factors such as changes in the regulatory burden facing listed companies will also be investigated. The extent to which New Zealand has become a “price-taker” in an increasingly globalised market for high-level executive talent will be explored, for example by checking for changes in the nationality of our CEOs over the past two decades.

A final note on context. In or around 2014, the companies in our database paid out $91 million in total annual remuneration to their chief executives. That is just 1/700th of total sales revenues of those companies, which totalled $65.5 billion. It is just 1/50th of the increase last year in the net wealth of New Zealand’s 180 or so richest individuals and families, as assessed by the National Business Review in its annual Rich Lists – up from $55 billion to more than $59 billion over the past twelve months. So why worry? Why even be interested in what our CEOs get paid?

Well, it may be that such rhetorical questions do occur to company boards and their compensation committees when they make their annual upward adjustments to top pay. But we have here also found a link between CEO remuneration and the size of the much larger managerial pay-bill, which totalled $653 million – around 2.5 per cent of total sales, and 20 per cent of total profits. Those numbers are beginning to get interesting.

From a different perspective, we have noted that the near-doubling of CEO pay since 1995 contrasts with increases in real wages and salaries just one sixth as large, on average. That could be a legitimate source of concern, from the point of view of fairness and social stability. Evidence on the actual productivity of top management should therefore be high on the agenda for future research.

Key take-outs

  • The pay of CEOs of New Zealand companies has almost doubled over the past two decades.
  • Managerial bureaucracy has expanded noticeably over the same period.
  • There is no apparent systemic link between remuneration and productivity or firm profitability.
Tim Hazledine

Professor Tim Hazledine teaches in the Department of Economics at the University of Auckland Business School. His research focuses on problems of competitiveness and unemployment in small trading economies, and competition policy, most recently in the context of air travel markets.



The alchemy of leadership

Getting "authenticity" wrong can be dangerous for both leaders and followers.



Calling the island to you

Polynesian navigators offer powerful leadership lessons.



Board led-transformation

Strategic renewal requires a board ready to drive a vision for change.



21st-Century Governance

Social and environmental accountability is becoming the new benchmark for directors.



Articles from our previous print editions can be found in our article archive:




Millennials are uncertain, not 'entitled'

We have read millennials wrong, argues Mike Lee. They can be as hard-working as anyone. What they most need are clear work boundaries in an uncertain world.




Hiring in the age of social media

Organisations should consider embracing gamification to enhance graduate selection and recruitment.




Socialisation of daughters in women-led family businesses

The global rise of women-owned businesses may be altering the way family and business interacts.




Conflict in a family business

Failures in family-owned businesses can often be traced to friction between family members.