Why Government Can't Make Decisions Rationally
On the rare occasions when governments consider curbing their expenditure on some service — or, more likely, consider curbing the rate of increase in spending — they are invariably called upon to undertake discussions with relevant "stakeholders" and members of the community. This process is ostensibly undertaken to determine the views of those who are to be affected by the proposed decision, to gauge the mood of the electorate our political masters are sworn to serve.
Of course, if one were to undertake a proper scientific study into the attitudes of the community to some such proposal, one would solicit the views of a random sample of community members. The use of random sampling ensures that there is no inherent bias in the sampling method — although bias can still result from non-response. This means that the responses of the sampled participants can be used as an estimate for the larger community, with statistical techniques used to determine how closely the former can be expected to reflect the latter. Indeed, this is a standard scientific technique for inferring the views of a large population, without the necessity for a complete census.
Suffice to say, this is not how the government proceeds. Instead, the government solicits the views of the relevant "stakeholders," identified by the bureaucracy. Who are these stakeholders? Why, they are the people and groups that have a stake in the decision of course. And when we say "a stake," we must be a little selective. After all, government spending is paid for, at least in part, by the myriad of taxes that are stolen from virtually every person in the population. And since we do not wish to take a complete census of the population, we must therefore limit ourselves to only some stakeholders, or more precisely, to those stakeholders who have the biggest stake in the decision.
For more major decisions, in addition to contacting the large stakeholders, the government may also feel obliged to hold a more general community consultation. Again, the government does not appeal to the scientific methodology of random sampling in selecting those community members who are to participate in this process. For this method would mean that members of the community would be excluded from participation by the roll of the dice; a most undemocratic result. Instead, the government simply invites the public at large to make submissions or attend some form of community forum. This process is self-selective: the respondents are those who themselves choose to take the time to write a submission, attend a forum, or otherwise make their views heard.
The result of this sampling mechanism is that both the stakeholders who are contacted by the government and the members of the public who self-selectively take the time to participate in the process are those who have a large stake in the proposed decision. And since the costs of any particular government service are spread diffusely through the population, whereas the benefits are concentrated among a much smaller group of rent seekers, it is precisely these rent seekers who will be the participants in the government's consultation process.
This result is nothing more than a manifestation of rational ignorance and rational non-participation by those with a small stake in the decision. To these people, the value to be gained from potentially influencing the government is outweighed by the time, effort, and potential stress involved, plus the opportunity loss from forgoing other activities (such as leisure time or paid work).
For the elderly retiree who relies on the public bus service to go to the doctor's office or to maintain his social life, the implications of a cut to this service may be severe. But for the ordinary taxpayer who does not use this service, the net financial benefit of a cut to the service is minimal, and thus, the net benefit of attending community consultation forums to speak in favor of this cut — even if successful — may be outweighed by the effort and stress involved.
But even aside from the question of participation, we must also ask who is to have more impact in this process. The elderly retiree, who is virtually in tears over the implications to his lifestyle? Or the taxpayer, who calculates that the cut to the bus service will save him $27 a year on his tax bill?
The outcome of this consultation is not in doubt. The consultation process itself will be weighted towards the interest of the rent seekers, both in numbers, and in the magnitude of their claims. Indeed, it would be a brave taxpayer who dared incur the wrath of "the community" by staking his meager financial claim against the high stakes rent seeking of others at such a forum.
This is true for each and every service of government that involves the diffuse spreading of costs among taxpayers. In each case "the community" that is consulted is composed of the rent seekers who benefit from this service at taxpayer expense. And so, in all such cases, those in the government who are desirous of greater statism — and they are many — have themselves a neat little set of loaded dice. For they may argue that they have "the community" on their side; that "the community" has spoken; that all those "with an interest in the decision" have been consulted. They may argue that democracy demands that the government abide by the views of this community and that it would be an affront to the democratic process to ignore these views.
Indeed, if the government proceeds to cut spending despite this consultation process, it is vulnerable to the accusation that the process was mere window dressing, and that the spending cut was a fait accompli. This may indeed be correct, given that there is unlikely to be anything emerging from the consultation process that supports such a course of action. However, this is only half the picture. For the truth is that it is often the outcome of the consultation process itself that is the fait accompli, skewed by its very nature towards the interests of rent seekers, who have a higher stake in the specific service under consideration than the taxpayers who are robbed to pay the bills.
 This is in fact a realistic figure for such an example. Taking my own home town of Canberra (Australia) as an example, the total expenditure on the public bus service in 2006–07 was AU$88.433 million, which amounts to an average of AU$273 per person (although the cost is not levied equally). Thus, a ten percent cut in expenditure on this service — which would involve a sizeable cut to the service — would amount to a saving of approximately AU$27.30 per person.