Could it happen here...

I was fossicking on The Irish Economy site, mostly to follow-up on the news that Ireland plans to come off the IMF/EU life-support machine next month, and I came across a post by Trinity professor Philip Lane, 'Public-Private Wage Gaps: EU Evidence'. This in turn took me to the source paper, a European Commission Economics Paper, 'The gap between public and private wages: new evidence for the EU', which is summarised here and available in full as a pdf here.

Here's the key finding.

Start with the bottom line. Reading left to right, the first column, 'Total difference', is the percentage difference between wage rates in the public sector and the private sector across the whole European Union. On average wage rates are 10.5% higher in the public sector (or were, anyway, on these 2010 numbers). The second column is how much of this premium can be explained by a vector of the usual suspects - education levels, age, occupation, level in the managerial hierarchy and what have you. As it happens, 6.9% of the EU-wide public sector premium of 10.5% can be explained by these compositional effects. And that leaves the third column, the 'unexplained' part of the premium, which I'm going to interpret as the extra wages you get merely for being in the public sector.

I've highlighted Ireland in yellow. Enough has been said already about the fiscal indiscipline of Irish administrations during the Celtic Tiger days, so I won't belabour it, but I will just observe that the Irish were the most profligate in the entire EU for overpaying public sector staff (by 21.2%), edging out Cyprus (20.9%) and Luxembourg (20.4%). It's noticeable, too, that all the PIIGS showed the same pattern of showering largesse on the public sector - Portugal (PT, 11.9% premium), Ireland (IE, 21.2%, as we saw), Italy (IT, 10.5%), Greece (GR, 8.2%) and Spain (ES, 15.1%).

You might conclude that the fix is wage cuts (as the Irish have since done) or at least a prolonged wage pause in the public sector until the premium is eroded by increases in the private sector. As the authors note, though, it's not that straightforward. There are systematic patterns to the overpayments: as they say (p28), "although a positive wage gap is found for public sector workers, this is mainly
concentrated on lower-skilled workers, typically occupying lower job positions", which in turn means that "fiscal consolidation measures aiming at reducing the public wage bill may find difficult trade-offs between the efficiency and equity goals".

Two thoughts.

One, I suspect the wage premium is only part of the EU overpayment picture, and if the full compensation package of relatively cushy job security, relatively generous pension arrangements, and contractual pay increases based on tenure* rather than performance were included, the comparison would tip even more in favour of the public sector.

And two, of course, you wonder, could it happen here? My first guess is, possibly not. If we're "most like" the UK in our arrangements, then maybe not - the UK shows as paying people marginally less (-1.3%) in the public sector (though that excludes the notoriously good pension deal many UK public sector staff enjoy). And anytime I've been involved in employment decisions in the New Zealand public sector, there have typically been attempts made to do a genuine like-for-like comparability exercise with what the job would pay in the private sector.

Those appointments, though, tended to be at the more senior levels, and on the EU showing, that's not typically where the gravy train is. It's lower down.

So it's still an open question. Anyone know of any evidence?

*As an aside, my father, a lifelong Irish public servant, once tried to prevent one of those payments (an "increment", in Irish civil service jargon) to one of his non-performing staff. It nearly caused a constitutional crisis.

How do you kill this zombie?

On Monday evening the chairman of the Australian Prime Minister's business advisory group gave a speech, 'Working with government to drive economic growth and a thriving business sector' to the Committee for Economic Development of Australia, CEDA. And it included this:

"In this regard [the challenge of Australia becoming more internationally competitive], it may be timely to review Australia's competition laws. It is clear, in a global sense, we are lacking economies of scale and that Australian companies find it hard to acquire the necessary critical mass in a small domestic market without running up against trade practices issues. If we are not ultimately to become a branch economy, the opportunity for Australian companies to become national champions at home must be considered by re-balancing the interests of consumers and businesses. To do otherwise is to encourage companies to shift to more friendly domiciles, sell to foreigners, or, if all else fails, to close their doors".

Other folk - most memorably John Quiggin in his book, Zombie Economics: How dead ideas still walk among us - have written about zombie theories that, however misguided, mistaken, illogical or outdated, live on and on and on.

Of all the zombie theories, this "national champions" zombie is the one that I would most enjoy burying at a crossroads at midnight with a garland of garlic around its neck and a silver stake through its heart.

Is there a sliver of truth in it? Maybe. Has the sliver turned septic and infected the rest of the body of the argument with dangerously noxious toxins? Yes it has.

The whole thing, apart from the manifest self-serving of the corporates who subscribe to it, has a core implausibility. We're supposed to believe that the way to create more competitive companies is to give them a less competitive home market where they can rort their own fellow citizens without having to worry about the fundamental drivers of creating an attractive product offering. Give me a break.

And I'm pleased that at least some of the Australian business media are having none of it, as in 'PM's top adviser trapped in a time warp', an opinion piece from the Sydney Morning Herald.

What really worries me, though, and also worried the Herald columnist, is that this line of thinking is re-stirring just when the Australian government is embarked on a bottom-up review of Australia's competition law. I don't currently have any good feel for the politics of it - anyone closer to it might care to throw in a comment or two - and I don't know whether the pollies are minded to tighten the law to deal more harshly with the usual suspects (banks, energy companies, supermarkets) or to loosen it to enable this "critical mass" guff. If it's for loosening, then from a competitiveness point of view, Australia is going to score a spectacular own goal.

And I'm also worried that this Australian zombie carries a communicable disease, which our pollies could easily catch.

Why not here?

You may have seen, last week, that Blockbuster, the American chain of video rental shops, decided to close the last 300 still open: it's been widely covered (for example here on Bloomberg), with quite a few commentators saying it was about time (as in 'Blockbuster is closing all of its video rental stores. Good riddance', on Yahoo! News).

Blockbuster, which at one point had 9,000 stores, wasn't much loved. Various people have commented on its limited range (other than multiple copies of the latest big hits), dreary stores, poor service, and overpriced extras (like popcorn), and the few times I used them in the States, I felt the same way.

Quite a few people also blamed Blockbuster for decimating the specialist, independent video stores, much like Amazon and the other e-booksellers have got it in the neck for dealing to that interesting little bookshop you used to go to.

But it wasn't its general unloveliness that did Blockbuster in. It was, first of all, Netflix's video-by-mail service (like our Fatso), and, more recently, the digital streaming of videos. And you can see why: online video on demand is cheaper (no bricks and mortar for the supplier, no getting the car out to go to the video shop for the buyer), faster, and offers an easily searchable and larger catalogue. Those of us out in the 'long tail' (I like French films) can find and view what we like.

Or at least we can if we live in the US. Here in New Zealand, we're well behind the curve. Yes, there's some online stuff available from the free-to-air channels and (in a rather limited selection) from Sky. And yes, Quickflix has entered the market, but judging by its current offering it's either early days or (if this is all the range they're going to offer) not a compelling proposition.

I try not to see market failures all over the place: for the most part, markets work pretty well to give us what we want. But I'm at the least bemused by the current state of play of video on demand in New Zealand. When what is completely routine in overseas markets is still the rarity here, you begin to wonder if we haven't got some kind of roadblock preventing or delaying the local deployment of a proven and popular technology overseas.

What a terrific outcome

Yesterday's employment and unemployment numbers were awesome, and defied even the most determined begrudgers to undermine them, though Radio New Zealand did its best by simply ignoring them in its 7.00am news bulletin this morning, preferring more important national items like the cost of insurance for maraes and Winston Peters' sniping at our chance of a Security Council seat at the UN.

The ever dependable Brian Fallow covered the data well at the Herald. His piece, 'Economic upswing flows into jobs', in particular pointed out that the unemployment rate fell, even as the participation rate rose. In other words, despite more people opting to join the labour force, there were more than enough new jobs to go round and still see the numbers unemployed going down.

Some critics like to say, "Aha! But this doesn't count the underemployed!", such as people working fewer hours than they ideally would have liked. That's right. But even on that score the latest numbers show things turning for the better: the underemployment rate, as opposed to the unemployment rate, dropped from 4.4% last September to 4.2% this September (you've got to use annual comparisons because the quarterly underemployment data aren't seasonally adjusted).

Another good outcome was what happened to the 'NEET' rate. This is mainly relevant to younger people, as it's the 'Not in Employment, Education or Training' rate (unemployment rates make less sense as a measure, as many young people tend not to be in the labour force in the first place). On that measure, which I rate as one of the more important social indicators, it's again all good. The NEET rate for all 15-24 year olds dropped to 11.4% in September, from 12.1% in June (it's seasonally adjusted, so the quarterly comparison is kosher), and is markedly down on the 13.4% of a year ago.

Despite all this good news, there's still been some attempt to have a beat-up on what looks like a lowish rise in pay. It's true that the 1.6% rise in what's called 'the labour cost index' over the past year isn't a huge rise. But that understates what's actually happened to folks' actual earnings.

The labour cost index is essentially what's happened to the rate of pay for a particular job. It doesn't include increases people might have got for working longer hours, or merit or performance bonuses, or the impact of promotions, or of people moving from one job to a better paying one. For that, you need to know what's happened to 'average ordinary time hourly earnings'. That's up by a more respectable 2.6% over the past year - not dancing in the streets material, I know, but still handily ahead of inflation over the past year (1.4%).

It doesn't usually get a lot of coverage, but there's also a table included in the labour market data that shows how we're faring by international comparison (using a standardised definition of unemployment). Here it is.

We're doing pretty reasonably, though not outstandingly, by OECD standards. There are 34 OECD members in this graph, and we rank 13th. I suspect we'll improve our ranking: I've highlighted Australia in green (5.8% on this basis), and it's pretty clear that we're going to overtake them, as consensus forecasts have the Aussie unemployment rate rising and ours falling.

Germany shows up to advantage, 7th in this group, with 5.2% unemployment. You might think this reflects well on how Germany's conducted its affairs (both at a business and policy level). You might be surprised, then, to discover that in recent weeks some of the blogosphere's heavyweights have climbed into Germany, essentially saying that they're exported their problems to the rest of the Eurozone. If you're interested, you could start with 'The real problem with German macroeconomic policy', or with Martin Wolf's critical piece in the FT, 'Germany is a weight on the world'.

Form your own views, but if the question is, should Germany be more like (say) France, or should France be more like Germany, I know which camp I'm in. Germany's got an efficient labour market, and France doesn't, and the social consequences of France's poor policy are enormous. 

Last month I read in the French business paper Les Echos about the latest annual survey put out by a French organisation that promotes the employment of younger people - '47 % des jeunes diplômés en 2012 sont sans emploi un an après', it said, '47% of young graduates in 2012 are without a job a year later'. And of the 53% that were employed, 30% were on short-term contracts, because of the employment 'protection' legislation that makes employers reluctant to take on full-time permanent staff. 

'Structural reform of the labour market' isn't exactly a snappy electioneering phrase, but in France's case it would go a long way to tackling appalling levels of youth unemployment and making proper use of the talents of its qualified young people.

A new state-owned insurance company?

Let me say two things first, before asking a question about the new state-owned KiwiAssure insurance company that David Cunliffe proposed as a new Labour policy over the weekend.

First thing is, none of us was born yesterday, we recognise that politics is politics, and we know that you don't expect to see detailed economic analysis in a speech to a party conference of why a new policy was thought up. You may not see any economic basis at all for some policies, and a good deal of the rationale was more nationalistic than anything ("Labour will confront the challenges of an insurance industry that is no longer Kiwi-owned", with the aim of "local ownership that keeps profits here").

Second thing is, I'm very pro-competition. I spent more than 12 years at the Commerce Commission, where the Commerce Act required us "to promote competition in markets for the long-term benefit of consumers within New Zealand", and I was minded that way even without the crack of the legislative whip. So I generally look pretty kindly on ideas that aim to improve competition and consumer choice, keep suppliers efficient, and keep costs down for buyers, which are among the stated objectives of KiwiAssure.

Currently, according to Mr Cunliffe, service from the insurance industry "we know from painful experience has not met Canterbury’s needs", whereas "Just as KiwiBank gave us a customer-focused, low cost Kiwi-owned bank, KiwiAssure will give everyone a choice for better service, competitive premiums and local ownership that keeps profits here...And like KiwiBank, it will offer customers an alternative and raise the bar across the insurance industry".

I suppose what's missing for me, thus far, is this: where's the solid evidence that the insurance industry is indeed non-competitive and inefficient?

Now, I know that it's not going to be an easy judgement to make: assessing whether an industry is genuinely competitive, as opposed to collusive or sleepy, is always a tricky exercise. But so far I haven't seen any good evidence put on the table that would convince me there's a screamingly obvious competition problem.

Premiums going up (if they are going up, which they likely are after the Canterbury and Wellington earthquakes) doesn't cut it as evidence on its own. Premiums can go up (and down) for all sorts of genuine, market-driven reasons. You don't start state-owned avocado farms just because avocados have hit $3 in the veggie shop.

For another thing, there look to be quite a lot of players: I counted 96 companies with full licences on the Reserve Bank's register of licensed insurers, and if the Insurance Council can be said to represent the bigger ones, then there are 29 companies on the Council's membership list. The number of players isn't, obviously, a slam dunk indication that there is lots of competition, and if the data in this article, Doubts cast on KiwiAssure's influence, are correct, then there's certainly one quite large player (IAG, which includes AMI, NZI, and State) with some 40% market share. So it's not a case of 29 (or 96) equals fighting it out tooth and nail. It's possible there could be some sort of cosy or dozy oligopoly operating, but normally, though, you'd expect an industry with that number of participants to be workably competitive.

The existence of the whole insurance broker industry is also somewhat suggestive that there are, indeed, competing companies offering different customer propositions. There are other interpretations (insurance is so complex you need professional help through the process), but on its face it looks as if there are people running around getting you competitive quotes from alternative providers.

What I'd like to see, and I've said so before - 'Competition to the rescue - again' - is that, if people feel that an industry is not competing for their business and isn't giving them the choice, prices and services they could reasonably expect, there should be an express power given to the Commerce Commission to look into it. You'd think they have the power already, but they don't, or at least not in any straightforward sense.

So let's see the evidence first. The policies can come second.

I'll drink to that

Last weekend I was browsing Jancis Robinson's wine column at the Financial Times, and discovered two interesting things I hadn't known from her latest article, 'Value judgments: my favourite wines selling for under £10'. Incidentally if you like her column, or the FT's mainline financial and economic coverage,  it's well worth either registering for free with the FT to get a monthly quota of articles, or subscribing, to get the whole online shebang, which is what I've done.

The first thing I didn't know was that New Zealand hosts what Robinson calls "the leading wine-price comparison site", high praise from a world wine authority. It's called Wine-searcher, and it is indeed very good. I can already feel a couple of cases of Côtes du Rhone coming on.

The second thing I discovered, and this is where we veer back into the world of economics, was that Wine-searcher has apparently done a seriously large-scale study of price dispersion on the internet (250,000 US wine prices, 74,000 UK wine prices). I couldn't find the study itself on Wine-searcher's site, so I'm reliant on Robinson's summary of it, which was: "Wine-searcher’s analysis found that it was not uncommon in 2003 for US retailers to charge up to 50 per cent or even 100 per cent more than the middle price; now, thanks to greater transparency, this is virtually unheard of. According to Wine-searcher: “In the past it was relatively common for merchants to charge 25 per cent more than their competitors. Currently only 6.5 per cent of merchants in the US and 6.2 per cent in the UK try to charge 25 per cent or more over the median price whereas historically these percentages were 27.1 per cent and 12.7 per cent respectively".

Robinson comments that "there is less gouging today than there used to be", and I'm sure she's right. In theory, suppliers charging more like each other doesn't necessarily mean that the market's become more competitive: near-identical prices are as consistent with near-perfect collusion as they are with near-perfect competition. But it's rather unlikely that there has been global collusion amongst multiple retailers across many, many different wine prices, and much more likely that consumers are getting a better deal as increased transparency has reduced the capacity of sellers to quote exorbitantly out-of-line prices.

This was one of the things that was supposed to happen as the internet became more popular, but it's taken a while to arrive. When I went to the AEA meeting in Washington in January 2003, there was a session on price dispersion and price convergence over the internet, and (as I remember it) the gist was that, at that stage, online prices were, somewhat surprisingly, just as disperse as the bricks and mortar ones. This is consistent with Wine-searcher's finding that there was still widespread price dispersion at that time.

I've been trying to find studies of pricing of other goods and services over the internet, to see if the same pattern of reduced price dispersion and better deals for consumers holds true outside the world of wine. So far, it's been an uphill struggle. If anyone can point me to good research in this area, I'd appreciate hearing from you.

And the wholesale price of your internet plan will be...

Today the Commerce Commission released its final decision on the UBA service (the wholesale cost a would-be internet service provider will have to pay Chorus to use the electronic gizmos that handle internet data).

The Commission based its decision on what the same service costs in some countries overseas ("benchmarking"), in practice ending up with the Swedish price, though the Danish price also went into the mix. The Commission settled on $10.92 per customer per month: add on the $23.52 the would-be provider needs to pay for access to the copper lines to the customer's premises (the 'UCLL' service), and the total wholesale cost of the line (UCLL) and the gear (UBA) will come to $34.44 a month.

Originally, in its draft decision last December, the Commission had suggested a lower total cost ($32.45, made up of $23.52 for UCLL plus $8.93 for UBA). This had turned the government incandescent: it looked as if the government-funded fibre-based network wouldn't get the uptake the government wanted, because copper would be so comparatively cheap. In August, in an update paper, the Commission had corrected some errors, as I described here, and the total proposed price had risen to $33.43 (still $23.52 for UCLL, but now $9.91 for UBA).

Chorus has taken it hard, but today's decision seems to me to be a highly sensible outcome.

For one thing, it's lower than the $44.98 a month Chorus charges at the moment, as it should be. The previous way of setting the price ("retail minus") tends to be over-generous to an incumbent owner of infrastructure. If that wasn't known to investors in Chorus - who have been peeved that its share price has been whacked by these regulatory developments - well, it should have been, or mostly should have been.

For another thing, it's lower, but not massively lower, than the entry-level wholesale price ($37.50) that would-be internet service providers will pay to access the fibre-based internet network, again as it should be. The fibre-based service is faster, and the slower copper-based service ought to be cheaper to buy. So far the government is saying very little, but if I were the government at this point, I'd be tempted to declare victory and go home.

I'm also pleased to see that the Commission flagged away the flakey idea it floated in its August update paper, when, they say today [52 - I'm quoting paragraph numbers here and later], they "set out an approach that would potentially extend the plausible range beyond the observed benchmarks". As they note today, "A number of parties were critical of this approach", which is putting it mildly: it was statistically naff.

The Commission has given it away, though, I have to say, rather begrudgingly. "The Commission remains of the view [they say at 53] that inferring a larger plausible range is a conceptually valid exercise of the IPP [Initial Pricing Principle, i.e. this whole exercise of benchmarking using overseas prices as a first stab], particularly where there are a small number of benchmarks" - well, no it isn't, actually, as professional statisticians who submitted to them pointed out - before finishing up with, "However, as we have not ultimately applied that approach in this determination, we have not responded to these aspects of submissions". Face saving? Sure. Got to the right place in the end? Absolutely.

I'm also pleased to see that three more countries (Belgium, Greece, Switzerland) were used as a rough and ready cross-check on the price set by using just Denmark and Sweden, even though the three countries did not strictly meet the criteria for being used as formal benchmarks themselves. There's a lot to to be said for being roughly right than precisely wrong.

And I thought setting the price point higher rather than lower within the benchmark range was right, too.

It's true that as Covec said [227], "the [adverse] effects on consumers [from a higher rather than a lower price] are certain but the effects on incentives to invest and incentives to migrate from copper
are uncertain". The Commission's answer [228] was that "We recognise the greater uncertainty of benefits but believe these uncertainties need to be considered against the potential negative consequences of setting the price too low. This could harm competition in the longer-term due to a loss in dynamic efficiencies", and [231] "we accept in principle that the risk to dynamic efficiency of a low access price is asymmetric and that the balance of risk favours setting a price that errs on the high side". This has been a central theme of all the regulatory telco decisions, and I think it's on the right track.

Another good LEANZ event

We've been having a good run with recent LEANZ events in Auckland, and last night's was no exception: Ed Willis from law firm Webb Henderson gave us a fine presentation on 'Promoting quality regulation' which was well attended and stimulated a lot of active discussion.

Ed, who describes himself as possibly the only person in New Zealand who is "passionate" about regulation (and he might be right, but that's OK), went through some of the tests for what good regulation would look like, instancing both Treasury's checklist - which you can look up for yourself in this Treasury paper - and a somewhat similar and equally useful paper I hadn't come across before, from the UK's Department for Business Innovation and Skills, 'Principles for Economic Regulation'.
Ed instanced two regulatory schemes as practical examples - the Overseas Investment Act, and the recent telco policy review - and his main points (on my reading) were two.

One, people can get confused between criticising the quality of regulation and disagreeing with the social policy intent of it. There'd be a lot of people (well, me anyway) who'd regard much of the policy intent behind the Overseas Investment Act as xenophobic, protectionist, and inefficient, for example, though that doesn't necessarily mean that the poor devils trying to implement the damn thing aren't making a halfway decent job of what's been dumped on them.

Two, sometimes regulation is trying to serve multiple and possibly mutually inconsistent purposes at once, which describes the telco policy review in a nutshell. You're unlikely to be able to satisfy everyone - if you want cost-based (and therefore probably lower) pricing for the copper network, you're going to upset the folk who don't want cheaper copper-based internet services getting in the way of the national benefits to be had from moving to a fibre network.

Ed was also pretty hot about the importance of transparency and accountability in regulatory frameworks, and who could argue.

In the context of accountability, could I give a nod to the folks at the Commerce Commission who have been doing the odd post-decision review of merger decisions, to see if things actually panned out the way they had thought at decision time? I know, there are folks who argue that the methodology for doing this kind of exercise is not up to scratch, and especially not if you're going to try and do a cost-benefit analysis and put $ numbers on the value of decisions, but I think that even a qualitative scan, with hindsight, of whether you got the main trends and factors right, is absolutely a step in the right direction.

Finally, thanks to NERA for generously hosting last night's event.