The idealised case for the social usefulness of the economics profession is that, within their area of expertise, economists offer the community dispassionate advice on the choices open to it, identifying the costs and benefits associated with each option and never failing to remind us that, whatever choice we make, it comes at an opportunity cost.
Unfortunately, those economists who belong to the economic-rationalist faction often fall well short of this ideal. They don't offer dispassionate choices so much as passionately advocate one option over others, often exaggerating the likely benefits and brushing aside both the direct costs and the opportunity cost of their preferred choice - which they sanctify by labelling a "reform".
Economic rationalists are the missionaries of materialism. They often step outside their expertise, failing to warn pollies or public that their figuring has taken no account of non-efficiency considerations such as how the increased national income they promise will be distributed between rich and poor, and how the reform will affect family relationships, leisure, stress, and trust and other forms of social capital.
This is why I was struck last week to see a lawyer, Geoff Giudice, former chairman of the (now) Fair Work Commission, doing what I don't remember ever seeing an economist do: reminding the community that "reform" comes with costs.
Giudice argued that, after four major reform acts on industrial relations law in the past 20 years, calls for further change should be approached with caution.
"Reform has a cost. There are significant transaction costs [I'd say switching costs] associated with changes (and attempted changes) in labour legislation.
"There is the cost of consultation at many levels: for example, internal deliberations in various representative bodies, governments, community organisations and special advocacy groups," he said.
"Lobbying, including representations to government and opposition parties, is not cheap. There are usually public relations and advertising costs of various kinds.
"Then the cost of the parliamentary process itself, including legislative drafting, the production of the associated parliamentary materials and sitting time, needs to be considered."
Should laws actually be changed after all that, there were usually significant implementation costs. In recent memory large amounts of money had been spent on changing staffing and other public service arrangements in response to legislative change, he said.
"There are other implementation costs related to compliance. Public information campaigns and industry education are needed.
"Employers can incur staff training, legal and other consultant costs ... Significant amounts of management time can be directed to dealing with proposals for legislative change," he said.
The economic rationalists would no doubt reply that the continuing efficiency gain from their proposed reforms would soon dwarf these essentially once-off costs. But this doesn't justify their failure to acknowledge the costs of the reforms they advocate, the way they behave like high-pressure salesmen.
It doesn't justify the cases where reforms fail to deliver the promised benefits because of "unintended consequences" (which their oversimplified model didn't foresee).
How often do reforms fail to bring significant net benefits? Probably a lot more often than we realise. Reformers are guided much more by their preconceptions than by empirical evidence. They tend not to dwell on their failures, shifting the blame to the pollies' flawed execution.
Since we live in a democracy, however, it's inevitable that governments' execution will fall short of textbook purity. This being so, the gap between theory and practice is a sort of cost the reformers should take into their reckoning before assuring us we've nothing to lose.
And the rationalists' mentality that the need for reform is never-ending, that for governments not to be reforming something means they're not doing their job, that too much reform is never enough, creates an environment in which we get too much change.
The reform mania gives rise to plenty of failed attempts to get reforms through, too much pseudo-reform - change for the sake of change as politicians seek to look busy - and too much oscillating change as we try centralising everything, then, when that doesn't work, try decentralising.
Particularly in industrial relations, we have too much battling between labour and capital to get the law slanted in their favour, all under the cover of "reform".
Little wonder Giudice thinks the goal of any further change ought to be making the law more acceptable to both sides, thereby producing long-term stability and certainty in the legislative regime.
He says that "when legislative change is proposed, all of the steps along the way from policy formulation, drafting, public debate and so on, through to implementation in some cases and abandonment in others, can be a wasteful distraction which displaces more productive activities".
It has taken a lawyer to remind the rationalists that reform itself has an opportunity cost.