Are public servants overpaid on average?

In my previous post I commented on the excessive growth of senior Queensland public service positions over the last few years. Let us consider the hypothesis that this growth was related to public servants having significant influence over their own employment conditions, and being insulated from the market forces that keep private sector earnings broadly related to productivity. As the Newman government discovered at the 2015 election, public servants are a major voting bloc and have considerable political power. Hence, public servants could end up being overpaid relative to their private sector peers. Certainly, average earnings are higher in the public sector than in the private sector (chart below).


Of course, we need to compare like with like, and take account of the differing nature of jobs and worker characteristics between the sectors. The public service workforce contains a higher proportion of professionals, so you would expect higher average earnings in the public sector than in the private sector. Nonetheless, rigorous empirical studies, which adjust for differences in job and employee characteristics, show there still exists a public sector wage premium, by which public servants earn significantly more than comparable private sector workers. This phenomenon has been observed in many empirical studies across the world.

In a 2017 study Public-private sector wage differentials in Australia, published in Australia’s leading economics journal the Economic Record, eminent labour economist Sue Richardson and colleagues from Flinders University reported:

After controlling for observed characteristics and individual fixed effects, we show that on average workers in the public sector earn about 5.1 per cent more in hourly wages than those in the private sector. The wage premium is slightly higher for females than males. Using a panel data quantile regression model with fixed effects, we show that the positive wage effects of public sector employment are heterogeneous, with comparatively larger impact at the lower end of the wage distribution than at other parts.

That is, the study finds it is ordinary public servants who benefit most from the public sector wage premium. Senior public servants are more likely to be paid similar to what they would earn in the private sector, although there is still a small wage premium (2.4%) on average. The study also finds that, among senior public servants, women are the main beneficiaries of the wage premium, while among ordinary public servants it is men who are the main beneficiaries (p. 114):

…low-paid public sector jobs appear to be favouring post-school educated men and high-paid public sector jobs appear to be favouring women in general.

The findings by Sue Richardson and her colleagues suggest there could be some positive discrimination in favour of women in the hiring process for the senior public service. This would be controversial if true. Further research on this issue would be highly desirable.

In summary, the best Australian evidence, which is consistent with the international evidence, supports the hypothesis of a public service wage premium. We should ask what is responsible for this premium? Is it the political power of public servants, one aspect of which is the existence of strong public sector unions?

The US public choice school that came to prominence in the 1970s provocatively analysed bureaucracy as if public servants are self-interested empire builders, rather than dedicated servants of the public good. In the famous 1977 article The expanding public sector: Wagner squared, Nobel laureate James M. Buchanan and fellow public choice economist Gordon Tullock argued government tends to grow ever larger, and public servants more powerful and better compensated, because public servants vote. Bigger government is in the interests of public servants, so they vote for candidates who support it. This leads to bigger government and more public servants, who also vote for candidates who support bigger government. This is the Wagner squared hypothesis. (Adolph Wagner was the nineteenth century German economist who proposed a law of expanding state activity.)

In my view, the Wagner squared hypothesis is too pessimistic—it came before the Thatcher and Reagan revolutions, and Australia’s own micro-economic reforms which all suggested the situation is not completely hopeless. But it reminds us that there are significant non-market factors that influence public service salaries. When negotiating future enterprise bargaining agreements, governments should keep in mind that, on average, their public servants are overpaid relative to their private sector peers.

Posted in Labour market, Uncategorized | Tagged , , , , , | Leave a comment

My comments on excessive growth in senior public service jobs in today’s Courier-Mail

Steven Scott from the Courier-Mail has done a terrific job of revealing the huge growth in the senior echelons of the Queensland public service since the change of state government in early 2015 (see Palaszczuk Government’s massive senior public servant hiring binge). Over the last three years, the number of senior public servants, at senior officer and senior executive service levels, has increased by around 30% (see my charts below which split out SO and SES levels).


Steven quoted me on the excessive growth in senior public service positions (on p. 7 of the Courier-Mail):

Former federal treasury official Gene Tunny said the rate of increase at the most senior levels of the public service appeared “excessive”, even when taking account of the extra senior managers needed to oversee frontline staff.

Mr Tunny, who runs Adept Economics, said the Government needed to look at ways to boost efficiency and should show how the extra numbers of staff were justified.

The state government claims that three-quarters of the additional senior positions comprise “doctors, nurses, health practitioners and other positions” in Queensland Health, but AMA President Michael Gannon, who is also quoted in the Courier-Mail article, suspects the additional senior Queensland Health positions are mostly in managerial rather than front-line roles.

One way to check whether it’s largely extra doctors and nurses or bureaucrats would be for the government to provide a breakdown of all the new Queensland Health SES and Senior Officer positions by specific occupation, and not just for a broad group such as professionals. Such disclosure does not have to identify individuals and raises no real privacy issues. (Even publication of the data at the ANZSCO Unit Group or Minor Group level would be revealing.) I know these data exist, because I extensively interrogated the public service database while working for the Beattie government’s Employment Taskforce in the early 2000s (see the public service database file specifications).

The government could provide much richer data on the public service than the limited cross-tabulations it currently provides. Also, the government should release a KPMG report it commissioned into recent public service growth. Much greater transparency on the Queensland public service is urgently needed.


Posted in Labour market, Uncategorized | Tagged , , , , , , , | 2 Comments

Titans threat to abandon Cbus confirms governments have wasted money on super stadiums

Based on the Courier-Mail report of a threat from the Gold Coast Titans to abandon the Cbus Super Stadium, over allegedly high charges imposed by Stadiums Queensland, the huge investments successive state governments have made in super stadiums (e.g. Suncorp, Cbus and now Townsville) may have been unnecessary. Hundreds of millions of dollars that would have been better spent on health or education priorities have been misdirected. The clubs can actually make do with smaller venues. The Courier-Mail reports:

The Courier-Mail has learnt that in a tense round of high-level negotiations, the Titans issued the explosive threat to sell home games to other venues both within Queensland and interstate, leaving Cbus Super Stadium as a $160 million white elephant…Titans officials got a taste of life away from the yoke of Stadiums Queensland when they went on the road to Toowoomba and Gladstone recently and were thrilled with the operational ease and commercial success of hosting those matches.

So the clubs don’t need the super stadiums after all! That would save the state government over $50 million in grants each year (see my Stadiums Qld post from yesterday). However, I expect the Queensland Government will find a more footy-friendly Stadiums Queensland board and it will reach a new deal with footy clubs so the government doesn’t suffer the political embarrassment of a $160 million white elephant. That, of course, would be a demonstration of the sunk cost fallacy. The government is losing money via Stadiums Queensland every year and it would be doing taxpayers a favour if it closed down the most uneconomic of its stadiums and sold them to the private sector to redevelop.

Incidentally, I had several excellent comments on my Stadiums Queensland post yesterday. Regular reader Brad suggests the governing bodies (e.g. NRL, AFL) could make a greater contribution:

One area of concern here is that the stadiums should be charging the national leagues much more money to recover costs. Each of the different leagues typically collect approximately 80% of their revenue from TV rights but those funds are not distributed to the clubs. Therefore, the leagues are getting rich while the government pays for the stadiums as the clubs have no money. The state government funding of stadiums enables the leagues to keep their money rather than them paying full cost recovery for the use of the stadiums.

Regular reader Jim noted it was a timely post:

…given the additions to the stocks of loss-making sporting assets on the back of the Commonwealth Games.

And Lateral Economics CEO Nicholas Gruen asked:

What is it with stadiums? All down the Eastern seaboard governments of both political persuasions seem to have lost their mind.

Finally, on the poor economics of the Townsville Super Stadium, see this excellent post from my old friend and former Treasury colleague Joe Branigan:

Townsville Super Stadium guest post by Joe Branigan

Posted in Gold Coast, Uncategorized | Tagged , , , , , , , | 6 Comments

Stadiums Qld is a financial drain

The Sunday Mail yesterday hinted the Queensland government may replace the current board of Stadiums Queensland, given that the allegedly high charges for the use of its stadiums, such as Suncorp, are causing financial issues for football clubs. I must say I have some sympathy for the current Stadiums Queensland board, which appears to be making the best of a bad situation. They have been put in charge of a loss making government-owned business with no prospect of ever making a genuine profit. The Sunday Mail reported yesterday that:

Stadiums Queensland says taxpayers should not foot the bill for costs associated with hosting sporting events when the Government had made considerable investment in construction of the facilities. Levy prices were fixed by TransLink and Queensland Police.

“Under this model, the hirer receives the majority of game-day revenue such as ticketing, signage, sponsorship and this means in Queensland, unlike many other states, our clubs have a greater potential to derive revenue from their events,’’ a spokesman said.

Stadiums Queensland made some excellent points there. Consider the massive capital costs of its operations. Its financial accounts from its Annual-Report 2016-17 reveal depreciation and amortisation account for 49% of its expenses from current operations, $59.3 million out of total expenses of $121.5 million (see figure below).


Sporting clubs and event promoters which use the stadiums do not even account for half of Stadiums Queensland’s revenue. The state government provided Stadiums Queensland with $51 million in grants in 2016-17 or 56% of its revenue (see figure below).


The financial statements for Stadiums Queensland reveal a 2016-17 operating loss on continuing operations of $28.7 million, but, thanks to some clever accounting, it reported net comprehensive income of $44.5 million. That clever accounting involved an increase in the asset revaluation surplus of $73 million, i.e. it was due to unrealised capital gains. The bulk of these capital gains were not on the underlying land on which the stadiums are sitting. In 2016-17, that land only appreciated by $3 million (see p. 50 of the annual report). The bulk of it was a $70 million revaluation of “improvements”, i.e. a revaluation of the stadium assets themselves, using a replacement cost methodology. The financial statements note (on p. 47):

The valuations have been determined using a cost approach (i.e. a modern/current replacement cost) due to there being no active market for such specialised facilities.

The reason there is no active market is because the stadiums are loss-making propositions. No one would buy them unless they could redevelop the land they are sitting on. Hence, although it complies with accounting standards, from an economic perspective it is misleading for Stadiums Queensland to include the revaluation of the physical assets in its comprehensive income statement. There is no way Stadiums Queensland could ever realise the capital gains it is relying on to pretend it is profitable.

The Sunday Mail claims it has evidence that Stadiums Queensland actually makes money for the state government:

Confidential financial records obtained by The Sunday Mail reveal millions of sporting dollars are pouring into Government coffers through the Stadium Queensland deals.

I’d like to see those documents, and unless they are made public there is no way of verifying whether the estimates are reasonable. The actual published financial accounts show Stadiums Queensland is a substantial cash drain on the state government. While the cash flow statement on p. 34 reveals a net increase in cash (and cash equivalents) of $25.1 million in 2016-17, it needs to be kept in mind the state government provided operating grants of $22.2 million and capital grants of $29.0 million that financial year.* So the net cash loss on Stadiums Queensland from the state government’s perspective was around $26 million. It goes without saying there is a high opportunity cost to subsidising Stadiums Queensland. Arguably, the money could be better spent on health or education priorities, for example.

*Technical note: the key to understanding how Stadiums Queensland can actually increase its cash at bank is to recognise that depreciation is a non-cash expense.

Posted in Industry policy, Infrastructure, Uncategorized | Tagged , , , , , | 13 Comments

South Brisbane & South Townsville top Qld in building approvals so far in 2017-18

In my last post, I considered the Queensland construction industry outlook. In today’s post, I take advantage of new data on building approvals at the small area level (ABS SA2 regions) for 2017-18 so far released by the ABS yesterday. The chart below of the top 20 Queensland small areas by the value of building approvals shows South Brisbane and South Townsville leading the state. Residential approvals dominate in South Brisbane, while non-residential approvals, no doubt mainly related to the $250 million North Queensland Stadium development, dominate in South Townsville.


Within the Brisbane metro region (see map below), we again see the familiar pattern of development being concentrated in the inner city and outer-lying areas, partly due to constraints on development such as heritage protection in many Brisbane suburbs, as discussed in Bradley Rogers’s 2013 QEW guest post Old Queenslanders in a new city.


In a recently published research paper Housing in Queensland, Queensland Productivity Commission (QPC) economists Matt Geck and Sean Mackay have noted that such regulations contribute to supply constraints on housing. They allude to the economic idiocy of such regulations when they note (on p. 33):

In large areas of Brisbane’s inner and middle suburbs, for example, character zoning is used to preserve the aesthetics of areas with clusters of well-located low-density housing built in 1946 or before. These clusters often surround some of Queensland’s highest-capacity public transport infrastructure, such as Park Road, Buranda, Dutton Park, Newmarket and Eagle Junction train stations.

That is, Brisbane City Council is restricting development in many of the suburbs best suited for population growth!

Incidentally, the QPC has an impressive capability to undertake economic research and it could be better utilised by the state government. It could, for example, analyse a potential switch from stamp duty to land tax, phased in over a 10-20 year period, as the ACT is doing and as QPC economists Geck and Mackay suggest may be desirable in Queensland (pp. 41-42 of their report).

I recall that, at the Australian Productivity Commission’s horizontal fiscal equalisation hearing in Brisbane in February, Under Treasurer Jim Murphy said Queensland Treasury had crunched the numbers on replacing stamp duty with land tax, but it couldn’t come up with a feasible model. According to the transcript on p. 596, Mr Murphy noted:

We’ve looked very hard at replacing all stamp duty with land tax, but it’s like other states, we’ve found it an incredibly difficult task to do that, especially as the impact on the state in the short-term.

Unfortunately, what happens in Treasury mostly stays in Treasury, so it’s hard for outsiders to judge whether Treasury’s analysis was adequate. Why not have the QPC investigate the issue and produce a public report for community debate?

Posted in Brisbane, Housing, Townsville, Uncategorized | Tagged , , , , , , , , , , , , , , , | 2 Comments

Queensland construction industry outlook

The construction industry is very important to the Queensland economy, and it employs nearly 10 percent of all workers (see Queensland Treasury’s Employment by Industry brief). One part of the broader industry, residential construction, has grown strongly in recent years, owing in part to the huge amount of apartment construction activity in Brisbane. But activity in residential construction has always been expected to fall back as projects were completed, and indeed it has been doing so. Private dwelling construction in Queensland in December quarter 2017 was 5.8% lower than in December quarter 2016. In contrast, non-residential construction activity is recovering nicely from its post-mining-investment-boom slump, with private non-residential construction increasing 11.6% over the same period. Growth in non-residential construction has therefore offset the adverse impact of the recent slowdown in dwelling construction on the state economy.

But what does the future hold? To gain some insight, we can examine building approvals data, the latest batch of which (updated with February data) were released last Wednesday by the ABS. Broadly speaking, as discussed below, the outlook is positive, based on recent approvals data and expected public sector capital works (e.g. Cross River Rail) and resources sector developments, possibly including the Adani mega mine (see this recent AFR article), although many observers remain doubtful it will ever proceed.

Non-residential construction

Non-residential building approvals have been at higher levels over the last couple of years after recovering from the trough in 2014-15 (chart below). This gives us reason to be confident about non-residential construction activity, although Queensland has not experienced the massive surge in non-residential approvals seen in NSW and Victoria.


A closer look at the non-residential building approvals data reveals a large part of the recent surge in NSW and Victoria was associated with buildings for education in the public sector. Also, Victoria has seen a doubling in the value of approvals for office buildings. Regarding Queensland, what struck me about the data is that we are disproportionately intending to invest in new retail and wholesale trade buildings, which strikes me as odd given recent trends in the sector.


Incidentally, Master Builders Queensland has a good summary of the outlook for non-residential construction in its 2018 Building Industry Outlook (p. 7):

2018 will see improvement in the demand for non-residential construction work with increased investment in a number of key industries that have been performing well.

The tourism industry will continue to perform well, requiring further investment in hotels and other short-term accommodation…Retail and wholesale trade buildings which is the largest source of demand for non-residential work, has already seen an increase in building approvals which bodes well for the coming year.

The office market will continue to work through an excess of supply and the industrial segment will see no growth.

Note that the non-residential approval figures I refer to above do not encompass engineering construction activity, such as that associated with roads, bridges, rail lines, and earthworks. But we can be confident regarding engineering construction activity due to increasing state government capital works spending (expected to increase from $7.9 billion in 2017-18 to $9.8 billion in 2018-19 according to the 2017-18 MYFER) and positive developments in the resources sector.

The Queensland Resources Council (QRC) provided me with a brief update on its sector earlier this week. Key points included:

  • According to IHS, Thursday’s coking coal price was USD 197.95—one year ago the price was USD 155.25
  • Four new coal mining leases were recently granted to Stanmore Coal securing 210 jobs
  • The Palaszczuk Government has awarded two preferred exploration tenders to two junior exploration companies, Metroof Minerals and Sojitz Coal; Metroof was awarded 86 square kilometres while Sojitz Coal received 45 square kilometres; both areas are within 60km south-east of Middlemount
  • Queensland based coking coal company Vitrinite, owners of several assets in the Bowen Basin, has signed an agreement with Japanese trading company Itochu Corporation to fast-track development of the coking coal Karin deposit

Developments such as these reinforce my confidence in the sector.

Residential construction

Regarding residential construction, Queensland’s new home approvals are on an upward trend (see chart below) and this will offset to some extent the decline in apartment construction. This was foreshadowed by Queensland Treasury last December in its 2017-18 Mid Year Fiscal and Economic Review (p. 8):

With the value of work in the pipeline now easing, it is expected that dwelling investment will fall in 2017-18 and 2018-19. However, a recent pick-up in approvals for houses, combined with a strengthening in population growth, suggests the decline in dwelling investment will likely be modest when compared with previous housing cycles.

Overall, Queensland residential building approvals appear to be at reasonable levels, while they have soared in NSW and Victoria.


Posted in Housing, Infrastructure, Uncategorized | Tagged , , , , , , , , | Leave a comment

Qld economy well positioned in 2nd quarter 2018, esp. given Commonwealth Games

We are now in the second quarter of 2018, and it’s a good time to review the economic outlook for Queensland. Job vacancy data released by the ABS last Thursday were very encouraging, and suggest the current economic upswing, encompassing a variety of sectors including health and aged care, education, tourism, and mining, will continue into this quarter. As Peter Martin pointed out in his article Best odds since the mining boom: 10 unemployed for each 3 vacancies, the ratio of unemployed persons to job vacancies is at a very low level nationally, so it’s a good time to be a job seeker, particularly in NSW and Victoria (see my chart below).


Peter Martin noted that Queensland was one of the worst states in which to search for a job. But job vacancies are actually surging in Queensland, too, being 23% higher in February 2018 than in February 2017. There is a good reason Queensland’s unemployed-to-vacancies ratio hasn’t fallen as low as in some other states and our unemployment rate remains higher (6.1% in Qld versus 4.9% in NSW, 5.7% in Victoria, and 5.5% nationwide). The reason is that Queensland has had a very strong recovery in workforce participation—i.e. an encouraged worker effect—as the state economy has woken up from its post-mining-investment-boom slumber (see chart below).

Participation rates

As you’re no doubt aware, Queensland’s economy is expected to receive a substantial positive shock early this quarter with the Commonwealth Games on the Gold Coast from 4-15 April. My colleague Nick Behrens, Director of QEAS, has a great post at his blog on the Games:

Is hosting the Commonwealth Games really worth it?

Nick has a good summary of the expected increase in visitors and expenditure associated with the Games:

More than 1.1 million visitors are expected in the lead up to, during and post the Games spending more than $870 million in Queensland.

Within this it is forecast that the Games themselves will attract approximately 672,000 visitors, spending $323 million (356,000 day trippers, spending $35 million; 265,000 domestic overnight visitors, spending $225 million 50,000 overseas visitors (including more than 6000 athletes and officials), spending $63 million).

The day trippers are probably mostly from SEQ so their expenditure is largely irrelevant to calculating an economic impact, because any Games-related expenditure reduces the amount they can spend elsewhere in the Queensland economy. Also, a reasonable share of the domestic overnight visitors will be from elsewhere in Queensland, and the same logic applies. I would guess the genuine Games-related additional spending in the Queensland economy in the 2nd quarter of 2018 would be around $250 million. This amounts to additional spending of around 0.3% of Queensland’s expected Gross State Product in the second quarter.  The spending wouldn’t all translate into additional GSP, because of leakages associated with imported inputs and also due to the crowding out of some other economic activity, but there is no doubt the positive shock will be substantial.

The state government is hoping for an ongoing rather than just a temporary economic boost from the Games, via greater tourism as a result of the international exposure, but I’m pretty skeptical about purported long-term economic impacts from sporting events. For example, the 2000 Sydney Olympic Games resulted in a surge in visitation for the period of the Games but no lasting impact (see this Conversation article Hosting the Olympics: Cash Cow or Money Pit? by two of Australia’s leading economic modellers, John Madden and James Giesecke).

Despite my skepticism about a Games-induced long-term boost to tourism, there is no doubt tourism will continue to contribute to the growth of Queensland’s economy. Generally, tourism has been growing at a good rate in Queensland (with international visitors up 4.3% in 2017), although less strongly than in NSW (up 7.4%) and Victoria (up 8.1%). For these figures and more, see:

TEQ’s 2017 International Tourism Snapshot

I noted Queensland’s potential to improve in international education in my previous post, and the same can be said for tourism more broadly. We need to remove the web of restrictive regulations that are constraining tourism investment and consider, for example, Brett Godfrey’s excellent suggestion to open up our national parks to eco-tourism opportunities (see his January 2018 Courier-Mail opinion piece).

Incidentally, based on Tourism Research Australia data released last week, domestic overnight travel (i.e. travel within Australia by Australians) is expanding at a healthy rate in Queensland, but at a significantly lower rate than in NSW and Victoria (see chart below). It’s time to reassess whether the $100 million or so provided by the state government each year to Tourism and Events Queensland could be better spent.


In summary, Queensland’s economy seems well positioned at the start of the second quarter in 2018. Note I will have a close look at the outlook for construction and mining in a post very soon.

Posted in Labour market, Uncategorized | Tagged , , , , , , , , | Leave a comment