Poker machines play a major role in redistributing income away from those with the least and towards those with the most. According to the Productivity Commission the social cost of poker machines is approximately $4.7 billion per year. Perversely, those who profit most from this most inequitable of income distribution devices often justify the existing regulatory regime on the basis that a small portion of the money lost in poker machines is spent on building sporting and community facilities in some areas. Just why the money lost by problem gamblers should be the primary source of revenue for improving sporting ovals is, however, typically left unsaid.

A number of inquiries have recommended reform to the way poker machines are regulated in Australia in recent years however the election of the minority Gillard Government in 2010 gave impetus to such calls for change with Senator Nick Xenophon, Andrew Wilkie and Senator Richard Di Natale all supporting specific reform proposals. Perhaps unsurprisingly, the gambling industry has responded more critically to proposals to reduce the money lost in poker machines than the representatives of the disadvantaged communities that the machines are alleged to contribute to. Indeed, the gambling industry has suggested that the introduction of reforms such as ‘mandatory pre-commitment’ technology or setting a maximum poker machine bet a $1 would cost between $2.5 and $5 billion to implement.

This paper argues that such estimates, when linked to the actual reforms being proposed, are so exaggerated as to be fanciful. The paper begins with a brief overview of the poker machine market in Australia. The second section of the paper outlines the methodology on which the $5 billion cost estimate is based. The third section provides a critique of the method used by supporters of the gambling industry to generate their $5 billion estimate. The final section provides new estimates of the likely cost of the proposed poker machine reforms and concludes that the likely cost of policy reform would be around one tenth of that being suggested by some opponents of gambling reform.

The paper also makes the observation that if the cost of poker machine reform was anything like the $5 billion being suggested by the Australian Hotels Association then the volume of software and hardware modifications involved would actually create tens of thousands of skilled jobs. But, that said, the main conclusion of the paper is that the costs of reform will be nothing like the $5 billion suggested by opponents of reform. The estimated reform costs from this paper are much smaller than those preferred by the Australian Hotels Association because they are based on the reforms under proposal rather than a hypothetical requirement to modify all poker machines simultaneously and immediately. Another major reason for the disparity is that the estimates provided below do not attribute the full replacement cost of decade old poker machines to the introduction of new regulation. That is, all commercial equipment has a commercial life and over the course of that life assets are typically ‘depreciated’ in value by their owners.

The $5 billion estimate, on the other hand, attributes the full cost of replacing a ten year old poker machine to the introduction of new policy. The estimates developed in this paper, on the other hand, make the more traditional assumption that all machines would eventually be replaced as they ended their commercial life and that the ‘cost’ of policy reform should only include the residual depreciated value of poker machines that have been replaced earlier than they might have otherwise been.