Championing social democracy not productivity

Not unlike the OECD, our Productivity Commission tends to lean left.  Not usually in some overtly partisan sense, but in a bias towards government solutions, a disinclination to focus on government failures as much as “market failures”, and a mentality that is often reluctant to look behind symptoms (which government action can sometimes paper over) to look at deeper causes and influences.

Sometimes the cheerleading for the left becomes more overt.   There was a streak of that evident in their climate change report a year or two back, but it seems particularly evident in their latest draft report out this morning.    Reflecting the change of government, the political complexion of key personnel of the Commission, the Commissioners –  while each individually capable –  appears to have shifted leftwards.

The Productivity Commission’s inquiries are into topics selected by the government of the day.  The current Minister of Finance has been keen on his “future of work” theme for years, dating back at least to when he became Labour’s Finance spokesman.  Now that he is Minister of Finance he is able to get the taxpayer to cover work in this area.  Here are the terms of reference for the current inquiry into “Technology disruption and the future of work”.

Apparently prompted by the government, the Commission appears to have begun releasing draft reports in stages.    This seems a useful step forward: potential readers or submitters might be faced with a series of 100 page drafts, not the single 500 page behemoths that the Commission often used to produce.  The draft report released last night (“Employment, labour markets and income”) is the second of five as part of this inquiry.

What it boils down to, amid various reasonable insights, is a push for a much bigger welfare state, allegedly in the cause of lifting average New Zealand productivity (and sustainable wages), without a shred of evidence or careful considered analysis connecting one to the other.    It is the sort of thing you might expect a political party to come out with –  the Labour Party conference, for example, is meeting shortly –  but not so much independent bureaucrats supposedly focused on productivity.

This, from the Commission’s press release, is the gist of what they are about.

flexicurity 1.png

I heard Sweet on the radio this morning playing up the contrast –  beloved of people championing flexicurity-  between “job security” and “income security”, claiming that New Zealand has the former (“a bad thing”) but not the latter.

The mental model that appears to be driving the Productivity Commission in this draft report is one in which we have an excessively rigid labour market, with people (and society) reluctant to face job change, and that this in turn is a big part of the reason why business investment has been low, and productivity growth has fallen far behind.  There is little or evidence adduced to support these claims, let alone the next leap that if only people were given more money more readily when they lost their jobs we’d be well on the way to solving our productivity failings.

Here is the Commission’s summary of the good and the bad, as they see it, of New Zealand’s labour markets (from p21 of their report)

flexicurity 2.png

It isn’t obvious that low labour productivity is particularly a labour market issue, but setting that to one side for now, I wouldn’t disagree with any of those bullet points.  But I’d look at the first group (“the good”) and be inclined then to suggest that this wasn’t the area to be looking if I was trying to do something about (a) economywide productivity growth or (b) adoption of new technologies by firms.  Our labour market looks pretty flexible and responsive. The Commission itself says so.

So why then the push for much higher welfare payments (call them “social insurance” if you like) to people who lose their jobs, less means-testing etc etc?  It can’t be economics –  facilitating ready movements of people from one job to another etc –  so it really has to politics; a view on a different set of income distribution arrangements.  That is stuff elections should be fought over.  But it simply isn’t that credible that the absence of these very general northern European approaches is causally connected to the failures of productivity here  (except perhaps in the reversed causation sense that richer and more productive countries may choose to be more generous).

The Commission is dead keen on us following the example of Denmark, the Netherlands, and Sweden.  These are all highly productive economies (which appear in the grouping of top tier countries –  northern Europe and the US – I often use here in productivity comparisons).  But it isn’t obvious that their labour markets function betters than ours does.  I’ve shown some comparisons re Denmark previously, when Grant Robertson in Opposition was touting the Danish “flexicurity” approach.   In that post, I concluded

I imagine that life on the unemployment benefit is a bit more pleasant in Denmark than in New Zealand, but it isn’t obvious that the Danish structure, as a package, is producing, over time, better outcomes than what we have here.  And their model is vastly more expensive, and more heavily regulated, consistent (of course) with Denmark’s position as the OECD country with the third largest share of government spending as a per cent of GDP (57 per cent).  New Zealand, by contrast, has total government spending of around 41 per cent of GDP

Perhaps more regulation and more spending was Robertson’s point.  I guess we have elections to debate such preferences, but it seems a stretch to believe it would be an approach that would make our labour market function better.  It isn’t obvious Denmark’s does.

But lets update the comparisons and extend them to include the Netherlands and Sweden as well.

First, take a look at the OECD’s indicators around employment protection legislation.  Recall the Commissioner claiming New Zealand has “job security” and suggesting we need to move further away from that.   Well, here are the comparisons.

The OECD indicators on Employment Protection Legislation
Scale from 0 (least restrictions) to 6 (most restrictions), last year available
Protection of permanent workers against individual and collective dismissals Protection of permanent workers against (individual) dismissal Specific requirements for collective dismissal Regulation on temporary forms of employment
Denmark 2013 2.32 2.10 2.88 1.79
Netherlands 2013 2.94 2.84 3.19 1.17
Sweden 2013 2.52 2.52 2.50 1.17
New Zealand 2013 1.01 1.41 0.00 0.92

New Zealand’s legislation around employment protection is more liberal on each measure than any of these flexicurity countries –  quite materially so in most cases.

Or unemployment rates, not just a one year snapshot but a glance back over this century to date.

flexicurity 3.png

Most of the time, most years, New Zealand’s unemployment rate has been lower than that in the median flexicurity country.  The differences aren’t always large, and aren’t even always same-signed, but it isn’t an obvious advert for the alternative model.

And what about employment rates?

flexicurity 4

Sweden has employment rates very similar to those in New Zealand, but taken together this group isn’t necessarily a great advert for an alternative income support model.  Of course, in richer and more productive countries more people can afford to work less etc, so I’m certainly not suggesting the whole difference is down to the presence/absence of flexicurity. Perhaps there is a political “income distribution” case to be made –  that’s one for political parties – but I’m struggling to see reasons why, evidence that, flexicurity offers potential labour market/productivity gains.

And to emphasise that flexicurity in these countries is just one component of a radically different approach to the role/size of government, here is a chart showing government spending.

flexicurity 5.png

The Commission even concedes (page 60) that materially higher income replacement rates when people become unemployed seems to be associated (in a cross-country relationship) with higher rates of (long-term) unemployment.  As they note, it isn’t an ironclad relationship –  there is always a lot of other stuff going on, which needs much more careful analysis to distinguish.   But why they would jeopardise one of the more impressive economic achievements of New Zealand this century (low averages rates of unemployment, especially long-term unemployment) isn’t clear.

Much of it comes down to this alleged “attitudes to technology” issue, even though the Commission makes no attempt at all to show that fears about technology or job displacement is somehow a major factor –  a factor at all for that matter –  in low rates of business investment in New Zealand.

They begin one section late in the report this way

flexicurity 6.png

So even though the Commission itself has concluded (reasonably, or so it appears to me) that “fears of mass job losses from automation [are] unsubstantiated” one public opinion poll is enough to suggest there is some widespread systematic structural problem.

One might well wonder whether (a) it was not ever thus (except perhaps in the hyper full-employment period of the 1950s and 60s, and (b) whether those fears are not being fed by people like the Minister of Finance.  Without evidence that any such fears are (a) much greater than usual, (b) causally connected to weak business investment in technology, and (perhaps) (c) evidence that such fears are lower in the flexicurity countries, it isn’t a great basis for proposing far-reaching policy change.

Following on from that extract we get this

Bredgaard and Daemmrich (2012, p. 2) described the Danish “flexicurity” system (Box 3.2) as a strategy for “economic competitiveness and sustainable national prosperity”.
Firms in Denmark gain competitive advantages from a mobile labour force and government funding of public services and infrastructure, while workers benefit from domestic employment opportunities and continuing training.

Well, perhaps, but I”m sure one can find champions for any country’s appraoch, but where is the systematic cross-country evidence, including relative to New Zealand (a country with lower average unemployment rates, lower long-term unemployment, higher employment).

And then there is this

flexicurity 7

But, as already noted, New Zealand not only has fewer job security protections than France –  the bad example cited here –  but fewer than those in Denmark, Netherlands, and Sweden.    And one might remind the Commission that correlation is not causation, especially when it isn’t supported by any independent argumentation to make the case that (a) flexicurity produces these poll results, and (b) more importantly, flexicurity increases the rate of uptake of new technologies across economies as a whole.   The Commission offers nothing on either point.

One could go on. The Commission notes that one “could” introduce “portable redundancy accounts” as, apparently they do in Austria. But makes no real case for doing so, and never seems to engage with (for example) the tax inefficiency (to the individual) of having various pots of money tied up in various places, all while (typically) having a mortage on the other side of the balance sheet.  They toy with ideas of mandatory redundancy, but again without any attempt to demonstrate a connection to productivity or business investment.  They worry about the ability of people who lose their jobs to service a mortgage, but never seem to adequately connect that concern to the fact that if there is a system that generates little long-term unemployment, most people are usually relatively readily able to find new jobs, and can self-insure (both formally, through mortgage protection insurance, and informally –  it isn’t common for both members of a couple to lose their jobs at once).  Mortgages in New Zealand are, of course, highly burdensome, but that is a reason to fix land supply and get the price of houses down, not to greatly enhance the welfare system.

Changing tack, I was also interested that the Commission did not touch on three other dimensions that might seem relevant to discussions around the sorts of schemes they propose:

  •  the fiscal automatic stabilisers in New Zealand tend to be quite muted.  That reflects the twin facts that our tax system isn’t highly progressive and that our unemployment benefit system is modest and pays a flat rate.     What the Commission proposes would strengthen the automatic stabilisers, but at the price of increasing the cyclical amplitude of cycles in the government’s budget balance.  There are pros and cons to such a change, but they didn’t seem to be mentioned at all,
  • the Commission rather overdoes the point about social insurance and how different New Zealand and Australia are to the rest of the world, but there is one important dimension they didn’t touch on.  In other countries, the social security systems are typically partly funded by social security taxes on wages.  That means tax rates on wages are typically higher than those on capital income.    This is a relatively attractive feature, given that business investment (especially foreign investment) tends to be quite sensitive to expected after-tax returns (and people like Andrew Coleman and me have been making this point for years).    Even if we did not increase welfare payments to the unemployed there would be a good case to lower income tax rates and raise the lost revenue through a social security tax on labour incomes.  This wasn’t a dimension the Commission touched on and while, considered politically, that might not be surprising, it is quite a gap analytically.

In sum, there is no sign that the current Productivity Commissioners have any sort of robust defensible model for thinking about New Zealand’s long-running productivity failures. In particular, they show no sign of having thought hard about why firms operating here –  and who might operate here –  have proved so reluctant to invest more heavily over long periods of time.   There is no evidence offered that excessive rigidity in the labour market, or fears of workers, is any part of the issue at all.

And yet they jump to champion quite radical changes in our welfare system, even including near the end of the report a folksy politicised cartoon

flexicurity 8.png

The economic case just is not made.  Sure, it would be great to have a highly productive economy, but the Commission simply has not made a serious effort to demonstrate any sort of causal connection between their (apparent) personal political preferences around unemployment benefits/social insurance and any sort of plausible path to much better productivity outcomes in New Zealand.   (And here one might note that places like France and Belgium –  with quite restrictive labour laws, much more so than the flexicurity countries –  have similarly high average rates of labour productivity.)

If they want to champion such a model –  and reasonable people can debate the merits of some aspects of it in its own right – there is an election next year. Perhaps the Commissioners might consider standing for Parliament instead of using taxpayer resources to champion a different answer to inherently political questions.