To start, the ACT has one of the lowest gambling machine tax rates for clubs of comparable jurisdictions, as the ACT Taxation Review found in 2010. Tax revenue from gaming machines would need to increase by $20 million to reach the national average. The review recommended increasing tax, which was agreed by the government, but six years on, ACT tax rates are an effective average of 19.9 per cent of gaming machine revenue, while the Australian average is 29.9 per cent.

In light of the relatively generous tax regime the government’s gaming machine “reform package” that started in August 2015 raises a number of profound questions.

The changes introduced an open market trading scheme for poker machine authorisations, and a phased reduction in the number of machines in the territory. The intention to reduce the number of machines dates back to 2012, with a target set at 4000, then a per-capita ratio was to be held. But unless there is a much greater reduction in the number of machines, and their withdrawal from the suburbs, placing a cap on numbers or setting a per-capita ratio is at best tokenistic and will almost certainly have no effect on poker machine use or problem gambling.

By way of example, data published by the ACT Gambling and Racing Commission shows that in 2014-15, the Canberra Tradesmen’s Union Club operated 400 poker machines, with each generating an average of $22,125 in net revenue ($8.805 million in total). In 2015-16, while the number of machines in the club decreased to 347 (or by 13 per cent), its net revenue increased to $25,599 per machine ($8.883 million in total). In other words, the remaining machines just became more efficient at generating revenue.

Prima facie, a market trading scheme would appear to be a reasonable way of allowing the market to allocate licenses across operators and venues, subject of course to some regulatory clearances. In the ACT, there is a particular issue, however, in that the existing licenses were granted at nominal cost – a grant of privilege with considerable value and in effect a licence to print money. The value of this privilege can be calculated as the net present value of the cash streams from each machine over its life.

Our analysis shows that the 489 poker machines in the Canberra Labor Club group have the highest value, at an average of $346,889 per machine. For the Spanish Australian Club, each of its 14 machines has an average value of just $20,755. The following table shows that the eight large club groups in the ACT, generating 87 per cent of the net revenue, have machines valued at $282,301 each on average. For medium-sized clubs (100 to 200 machines), each machine is valued at $186,011, and for small clubs (fewer than than 100 machines), the value of each machine on average is $100,672.

Clubs are of course entitled to the varying revenue streams that determine the average value of their machines. A question that this analysis begs – in fact it might be described as the $64 million question – is how does, and indeed should, a market trading scheme deliver a windfall gain to the clubs which operate poker machines?

Enter the decision to allow poker machines in the casino. The perplexing decision to allow 200 machines in the casino – itself a reversal of a previous decision not to depart from the community gaming model – can now perhaps be explained. It creates a ready market for machines that would otherwise not exist. Clubs that are able to reduce their machines while making the remaining machines work harder will gain a windfall benefit. It is also possible for a large club to buy machines from smaller clubs, and store them for eventual sale to the casino.

In short, a cap on gaming machines, a market trading scheme, and the grant of a 200 machine licenses to the casino combine to deliver significant windfall benefit to gaming machine operators. Those benefiting will most certainly be the large clubs, including in particular those owned by or associated with the Labor Party and the CFMEU.

We do not believe there are sound policy grounds for allowing gaming machines in the casino. The government has not released a business case or a cost benefit analysis for abandoning the community model of gaming and adopting a market trading scheme for licenses granted for free.

It can be argued that a scheme to allocate new machines, or to reallocate the existing machines is needed as Canberra grows and the existing community clubs expand into new suburbs, or new clubs are formed. A market trading scheme, however, will result in the financial benefits of the reallocation of existing licenses accruing to the private operators and not the Canberra community.

In 2010, Victoria auctioned off 27,500 of its gaming machine licenses for a 10 year period from 2012, raising approximately $980 million, an average of $35,636 per machine. If the machine values achieved in the Victorian auction, which were lower than expected, were realised in an auction in the ACT, taxpayers would receive $7.1 million for 200 machines.

In an ACT auction, the price which potential gaming machine owners (the casino in this case) would be willing to pay for a license would be up to the net present value of their forecast of cash flows from operating the machine. If the operators expected just the average (and not the maximum) of what the large clubs earn in net revenue, the return to the clubs selling those machines would be $56.5 million. In principle, that revenue belongs to the ACT taxpayers.

If the government wished to reduce the number of gaming machines, it could use a voluntary surrender scheme, and if necessary, a compulsory acquisition scheme. The value of compensation to the clubs in each scheme could be easily determined for each of the machines surrendered or acquired.

The ACT Gambling and Racing Commission collects data for each of the machines being operated. It is well known that not all machines have equal patronage; some are preferred and the difference in revenue between the highest performing and lowest performing machine may be many-fold. As such, a financially responsible scheme could be designed that provides fair compensation for the machines surrendered or acquired, while ensuring that the net benefits (ie, revenue from auction less the acquisition cost) are properly returned to ACT taxpayers.

Subject to the model ultimately adopted by the ACT Labor/Greens government, the Labor and Labor affiliated clubs could be enriched by tens of millions of dollars. It should be a given, having regard to the Labor Party/CFMEU association with and ownership of poker machines, that the Labor/Greens government should because of its irreconcilable conflict of interest disqualify itself from any involvement in the regulation of gambling in the ACT. Alternatively, Labor and Labor affiliated clubs could give a commitment that they will surrender all of the machines required of them without recompense or compensation.

Jon Stanhope is professorial fellow at the Institute of Governance and Policy Analysis,
University of Canberra, and former Labor chief minister

Khalid Ahmed was executive director, policy coordination and development, for ACT Treasury. He is adjunct professor at Institute of Governance and Policy Analysis, and works in the private sector.