How to do labour market liberalisation
Given that WorkChoices-killed-the-Howard Government is already the received wisdom*, labour market liberalisation may be dead for a generation. Still, failure is instructive, so it is worth (if only as an intellectual exercise) considering what we can learn from WorkChoices about how to do (and not to do) labour market liberalisation.
(1) Don’t make it a surprise.
If you want to liberalise the labour market, let the electorate in on it. Don’t spring it on them. Major change they haven’t voted for is much, much harder to sell. And that’s entirely fair enough too.
(2) Create a package you can sell.
Which means something with a clear idea behind it that doesn’t get lost in interminable details.
(3) Don’t pretend there will be no losers.
Pareto optimal change is a nice idea for economists. In the real world, there will always be swings and roundabouts, even if only in the short term. Indeed, the economists’ notion of Pareto optimal improvement (a change whereby, even if the losers are fully compensated, still leaves a net gain) does imply actually compensating losers.
(4) Have a clear mechanism for helping losers
The Hawke Government got folk to wear increasing employment by reducing real wages through offering “social wage” improvement - government provision compensating folk for failure of their wages to rise in line with inflation. (Releasing resources wasted or blocked by labour market regulatory complexity was largely a no-goer because of union opposition, as the unions lived off such complexity.)
An ostentatious penalty regime within the package is not compensation for losers. It is a rather ham-fisted promise that you will scare employers into there being no losers. Moreover, you are buying back into the idea that labour market regulation is “special” and a mechanism for social protection. The point is to normalise labour law, not keep it as a separate playpen.
The problem with labour market regulation is it attempts to do by clumsy and inefficient mechanisms social protection that the tax-and-transfer system can do rather more efficiently. So, if one is going to liberalise the labour market, be absolutely upfront about delivering social protection better. After all, the lower one’s income, the more important continuity of income becomes. Measures such as raising the tax-free threshold, bottom-end family payments and earned income transfers would all show that one was taking people’s concerns seriously. (Besides, increased employment and wages would, over time, reduce the cost of such measures. Raising the tax-free threshold also reduces tax-transfer “poverty traps”.)
It will also blunt the main appeal of labour market regulation - to protect job incumbents. Which is the main problem with labour market regulation - it protects job incumbents at the cost of excluding those whose productivity or perceived risk level doesn’t make the bar set by the regulation. So, for example, labour market de-regulation in New Zealand lowered Maori unemployment rates far more than any other policy change ever had.
Even if the point is to punish the unions, nothing will do them more damage than regulatory simplicity. Then they really would have to work for a living, and their failure to hold their client base (the proportion of employees who are union members having fallen from 41% in 1990 to 20% in 2006, in the private sector from 31% to 15%) shows they are bad at that. (Which is why they have invested so much in guaranteeing political privileges.)
A serious enough surge in unemployment could be an opportunity to revisit the issue, provided it is cast in terms of “costly regulation makes things worse” and “social protection can be delivered better”.
* Though I think that the housing market also played a role, particularly in explaining major differences in swings by State.
(1) Don’t make it a surprise.
If you want to liberalise the labour market, let the electorate in on it. Don’t spring it on them. Major change they haven’t voted for is much, much harder to sell. And that’s entirely fair enough too.
(2) Create a package you can sell.
Which means something with a clear idea behind it that doesn’t get lost in interminable details.
(3) Don’t pretend there will be no losers.
Pareto optimal change is a nice idea for economists. In the real world, there will always be swings and roundabouts, even if only in the short term. Indeed, the economists’ notion of Pareto optimal improvement (a change whereby, even if the losers are fully compensated, still leaves a net gain) does imply actually compensating losers.
(4) Have a clear mechanism for helping losers
The Hawke Government got folk to wear increasing employment by reducing real wages through offering “social wage” improvement - government provision compensating folk for failure of their wages to rise in line with inflation. (Releasing resources wasted or blocked by labour market regulatory complexity was largely a no-goer because of union opposition, as the unions lived off such complexity.)
An ostentatious penalty regime within the package is not compensation for losers. It is a rather ham-fisted promise that you will scare employers into there being no losers. Moreover, you are buying back into the idea that labour market regulation is “special” and a mechanism for social protection. The point is to normalise labour law, not keep it as a separate playpen.
The problem with labour market regulation is it attempts to do by clumsy and inefficient mechanisms social protection that the tax-and-transfer system can do rather more efficiently. So, if one is going to liberalise the labour market, be absolutely upfront about delivering social protection better. After all, the lower one’s income, the more important continuity of income becomes. Measures such as raising the tax-free threshold, bottom-end family payments and earned income transfers would all show that one was taking people’s concerns seriously. (Besides, increased employment and wages would, over time, reduce the cost of such measures. Raising the tax-free threshold also reduces tax-transfer “poverty traps”.)
It will also blunt the main appeal of labour market regulation - to protect job incumbents. Which is the main problem with labour market regulation - it protects job incumbents at the cost of excluding those whose productivity or perceived risk level doesn’t make the bar set by the regulation. So, for example, labour market de-regulation in New Zealand lowered Maori unemployment rates far more than any other policy change ever had.
Even if the point is to punish the unions, nothing will do them more damage than regulatory simplicity. Then they really would have to work for a living, and their failure to hold their client base (the proportion of employees who are union members having fallen from 41% in 1990 to 20% in 2006, in the private sector from 31% to 15%) shows they are bad at that. (Which is why they have invested so much in guaranteeing political privileges.)
A serious enough surge in unemployment could be an opportunity to revisit the issue, provided it is cast in terms of “costly regulation makes things worse” and “social protection can be delivered better”.
* Though I think that the housing market also played a role, particularly in explaining major differences in swings by State.