Friday 27 March 2020

Lessons for later

There will be lots of economic policy lessons from covid-19. One of them, I hope, will be about the coordination of fiscal and monetary policy.

Before the virus, there were many people saying that monetary policy had already been loosened so much that it left central banks too little room for further support if the proverbial encountered the wind redistribution device. And they were right.

Have a look at this simple and I'd say uncontroversial macroeconomic policy schema. There are four states of the world, too high/low inflation crossed with too high/low unemployment. Some folks might prefer an 'output gap' to 'unemployment' but same diff.

What's 'too high' or 'too low' inflation? Judgement call, but there's enough of a consensus these days around 'materially above/below 2%' as a good enough rule of thumb. And 'too high' or 'too low' unemployment? Unemployment is unwelcome at any substantial level, but for these purposes let's call it 'materially above/below the level that would get inflation accelerating' (the 'NAIRU' in the trade). Nobody has a very tight grasp on that level, but (according to Figure 5.3 down the back of the RBNZ's most recent Monetary Policy Statement in February), it's somewhere in the 4% to 4.75% area. So let's say 'clearly above 4.75%/clearly below 4.0%'.


In principle, in two states of the world (the green boxes) fiscal and monetary policy ought to have been pulling together. In the top right quadrant, anything the RBNZ did to boost inflation would likely do the real economy some good, and ditto for fiscal policy helping to increase inflationary pressures.

Did any of that happen? No. In practice the official cash rate got all the way down to 1.0% before fiscal policy belatedly came to the party by way of the $12 billion infrastructure spending plan in the December 11 Half-Year Economic and Fiscal Update. Here's the stance of fiscal policy, from that HYEFU.


During the whole of the 2012 through to 2017 period fiscal policy was actually contractionary - not expansionary, as it should have been, since unemployment was too high through all that period as this chart from the RBNZ shows (again from February's Statement).


I've painted things rather black and white, and there are some nuances being left out. One is that I can see some of what contractionary policy was trying to achieve which, in part, was to restock the ammo after the splurge on anti-GFC and post-earthquake support. And another is that whether the OCR was 5% or 3% or 1% at the onset of the covid-19 out breakout has become somewhat moot, since monetary policy would have seen the OCR cut to its present 0.25% and unconventional monetary tactics deployed either way. And I'm conscious that we're far from the worst in the world at this: the eurozone, for example, pushed monetary policy even further than we did (into negative interest rate territory and into tactics like quantitative easing) and were even more feeble on the fiscal front.

All that said, there has to be a better way. Pushing one setting of policy to Full Steam Ahead and leaving the other on Mild Astern makes no sense in periods when they should be coordinated (like most of the past decade). When we get out of this, we need No 1 The Terrace and No 2 The Terrace to get their act together.

Tuesday 24 March 2020

Patrolling a fine line

There is a growing disquiet - particularly in the US, but also elsewhere - that competition policy has got too soft on mergers, and has been permitting anti-competitive increases in market concentration. By coincidence the latest in ComCom's excellent series of telco market monitoring reports included some data on the international costs of mobile plans and broadband. It showed just how acute the problem has become in the US, which is well entrenched on the expensive right hand side of both graphs.


As Thomas Philippon said (pp5-6) in his recent book The Great Reversal: How America Gave Up on Free Markets, "In most advanced economies, consumers pay around [US]$35 for broadband internet connections. In the US, they pay almost double. How on earth did that happen? How did the US, where the internet was "invented" and where access was cheap in the 1990s, become such a laggard, overcharging households for a rather basic service?"

His answer, which feels right, is that incumbents' market power to charge more and/or give less has increased as some markets have become overconcentrated. "First, the entry rate of new businesses has declined. Businesses are now older and face fewer new competitors reach year. This has led to concentration from the bottom up", and he says that "lobbying and [new-entrant-unfriendly] regulations explain much of the decline in entry rates". "Second, agencies and judges have allowed more frequent mergers among large businesses. This has led to concentration from the top down. Together, they account for the rise in concentration that we have observed" (all from p96).

However true that may be, there is a risk that the backlash goes too far the other way and puts in jeopardy mergers that would make pro-competitive sense, despite the first-blush look of yet another increasingly concentrated market. While the latest big mergers won't make any of the "you've gone soft" critics happy, they look to have been the right call.

In the US, the example is the merger of the number three and number four mobile telcos, T-Mobile and Sprint. You might think, looking at the left hand panel in the graph above, that allowing further concentration in the already expensive US mobile market would be a difficult proposition to justify. And yet: the argument for it is that a bulked up three-plus-four will be a more effective competitor to the two big incumbents (AT&T and Verizon) than three and four separately would have been. There's a logic to that: it was raised, for example, in one of the first merger clearances I was involved in (Telstra/Clear in 2001) though in the end the merger was cleared along traditional ongoing competitive constraint from incumbents/new entrants lines. In the US, both the Federal Communication Commission and the Department of Justice were prepared to go along. Individual states weren't best pleased, but the last of the them standing, California, has also folded its hand after negotiating some state-specific goodies.

In Australia, it's similar. The ACCC had opposed the TPG-Vodafone merger: the ACCC reckoned that left to its own devices, TPG would have rolled out a fourth mobile network to compete with the big three (Telstra, Optus and Vodafone itself). The case in the end turned out to be less a contest of economic merger perspectives and more one of disputed facts. The court disagreed with the ACCC's view of the world: at [34] the judge said that "The Court has been left in no relevant uncertainty, after reviewing the evidence, as to the future of the retail mobile market which will not involve Mr Teoh [TPG's executive chairman and CEO] or TPG entering the Australian retail mobile market in the next five years" (full judgement here). Earlier this month the ACCC flagged away appealing.

Incidentally, as any of us who have to try and guess where competition litigation will go will agree, you have to tip your hat to Tony Boyd, the Chanticleer columnist at the Australian Financial Review, who in a (possibly paywalled) piece 'ACCC misreads mobile market' back in May 2019 absolutely nailed it: "ACCC chairman Rod Sims is headed for what is almost certainly an embarrassing defeat in court ... Rod Sims is kidding himself if he thinks the Australian Competition and Consumer Commission can win". He got it bang on: Justice Middleton could have saved himself most of his 899-paragraph judgement if he'd just copied and pasted Boyd's conclusion that "The problem with the assumption that TPG’s executive chairman David Teoh will spend billions of dollars building a new network is that it is completely without foundation".

So even against an appropriately heightened sense of scepticism about the merits of even further mergers, ones will still pop up there may be no loss of competition compared with the no-merger counterfactual (as in TPG-Vodafone) or there is the procompetitive emergence of a more effective post-merger competitor (T-Mobile-Sprint). Whether any of these arguments apply to the other behemoth out there - if it's survived the recent market havoc, the proposed merger of  the global number two and number three insurance brokers, Aon and Willis Towers Watson, to create a new number one - will remain to be seen. There look to be a good deal of potential back-office efficiencies, but whether they, and the claimed capability synergies, make up for the potential reduction in competition is debatable.

Finally, the latest ComCom telco report got me digging in the archive. Here are a couple of graphs from the very first telco monitoring report, published back in 2008.



Nor pretty at all back then, was it? The improvement since then, especially on the mobile side, has been enormous. And a lot of it goes to prove Philippon's points: for a good competitive outcome you can't beat new entry (particularly 2degrees) and entry-facilitating regulation (a whole swathe of things, from early 'ladder of investment' ideas like wholesale access through to mobile termination regulation and local loop unbundling).

Wednesday 11 March 2020

One Rule to bring them all and in the darkness bind them

"Chatham House rules", said the invite, though as Chatham House itself says, "There is only one Rule", namely
When a meeting, or part thereof, is held under the Chatham House Rule, participants are free to use the information received, but neither the identity nor the affiliation of the speaker(s), nor that of any other participant, may be revealed.
So let's just say I've been at a forum where two items got discussed - whether the institutional set-up of the Commerce Commission is all that it might be, and whether our Court of Appeal threw a lit firework through the windows of the Commerce Act.

The first bit was partly a push-back response to the New Zealand Initiative's 2018 Who Guards the Guards? - Regulatory Governance in New Zealand (media release here, fuller version here), but even if the Initiative had never raised it, it's a good idea to talk about, and MBIE or ComCom itself also ought to be kicking the tyres from time to time and checking that processes and policies are fit for purpose.

The Initiative argued that "the separation of board and executive functions at regulators like the FMA [Financial Markets Authority] contributes to greater accountability and to greater levels of expertise.  Commission models like the one in place at the Commerce Commission fared worse". In other words, in the Initiative's preferred state of the world, Commissioners would set the general strategic direction of the place, and hold the CEO accountable for delivery, in the way that a corporate board would, but the CEO and the staff would make the actual decisions on mergers, trade practices, and sector regulation.

I didn't find the Initiative's conclusions persuasive at the time about ComCom, and after this latest forum I remain unconvinced. It's true that the Initiative were onto a worthwhile topic, and also true that the data they collected from survey respondents on a wide range of regulators (which you can download for yourself here) were fascinating. But if you unpack ComCom's performance, you find some interesting patterns.

Its worst performance on the 23 criteria the Initiative canvassed came from "The ComCom understands the commercial realities facing your industry", where 57.9% strongly disagreed or disagreed. On its face, you can see why the Initiative would then call for "broadening the skills set of the commissioner/board members of the commission to include members with industry expertise".

But hang on: when asked whether "The leaders of the ComCom are skilled, knowledgeable and well-respected by businesses in your industry", they came out well, with only 18.9% strongly disagreeing or disagreeing. The staff at ComCom, ditto, with only 19.4%. That's hard to square with the picture of clueless wallies.

The simplest explanation of these two contrasting pictures is that the "you don't understand our industry" response is special pleading. "If only you understood airlines/electricity/shipping/insert professional service here, you'd realise why we need to be able to agree on prices/set capacity/restrict entry". And you might start thinking dark thoughts about regulatory capture if the likes of a ComCom ever got too-high marks for helpfully going along with old Spanish practices.

The next worst ComCom score, by the way, was on the question, "You are not hindered or deterred from taking action to improve the profitability of your business by any lack of predictability in the ComCom's decision-making", with a 54.1% thumbs down. Noting that this doesn't really speak to the issue of form of governance - a structure where the staff make the calls could be just as unpredictable as one where Commissioners do - my guess is that the low score says more about the state of New Zealand competition law (and in particular around the state of the law on abuse of market power)  than about ComCom structure or process.

At the forum I was at, the point was made that maybe we ought to have a comprehensive review of our competition law, as the Aussies did in 1993 (the Hillmer review) and again in 2015 (the Harper review). I tend to agree, though clearer or better legislation won't be the complete silver bullet: drawing the appropriate line between vigorous competition from big players and anti-competitive use of market power remains problematic in every jurisdiction, irrespective of the wording of the law. But taking whatever steps we can to clear the air (and adopting Australia's "effects test" would be a good start) would go some way to address the uncertainty the Initiative's respondents were voicing.

Finally, the other major area the Initiative survey identified as a concern was "The ComCom is readily held to account for the quality of their work (including any mistakes) by its responsible government department, minister, or some other effective external accountability mechanism", where 44.1% were unhappy. This seems to be a widespread concern across all 24 regulators asked about: the average answer is 43.4%, so ComCom's rating is entirely typical.

My personal feeling is that in ComCom's case the response may reflect the lack of visibility of the accountability mechanisms, which are stronger than people may realise. If, for example, you had to operate within and report on the tightly hypothecated budget buckets that MBIE makes the Commission use, I'd bet you wouldn't like it. And while Select Committee oversight tends to be for policy trainspotters rather than for the wider community, Committee members can be robust in their attitudes to the Commission (as I noticed for myself when I turned up to argue the case for the Commission being able to initiate market studies). Again, the state of the law may also be a factor: ComCom in its everyday business isn't subject to "merit reviews" through the courts. It can be challenged on the law, or on due process, not on substance. You can understand why people might feel frustrated at not being able to relitigate the facts.

While I'm sympathetic to the Initiative's accountability concerns, I'm not sure that the corporate board model is the right one in this quasi-judicial context, a point also made at this forum. I wouldn't want a Supreme Court sitting back and managing the clerks' budgets, while the (nameless?) clerks make the decisions.

The second part of the forum dealt with the possibility that the courts may have further muddied the legal waters, by calling into question how the Commerce Act is supposed to work. The case  is the Court of Appeal's 2018 NZME/Fairfax judgement, and the argument is about how you decide whether something is on balance a good idea, after taking all the benefits and detriments into account.  Is the test, the country as a whole will be better off (a "total welfare" standard)? Or is it, consumers will be better off (a "consumer welfare" standard)? Or something in the middle where the country's better off and at least some consumers see some of the benefits (a "modified total welfare" standard)?

At the time (including in my own coverage) the big issue in the NZME/Fairfax case was seen as what things can properly be counted as benefits or detriments, with a big focus on whether ComCom was entitled to count loss of media plurality as a detriment (it was). Nobody paid huge mind to the bits of the judgement dealing with 'The Act's objectives' in [42] to [53], with Australian precedents in [66] to [68], or the punchline in [75]. Essentially, the judgement said, "Look, ComCom, you've always used a total welfare standard. You seem to think the courts required you to. Don't know about that - we in this court certainly didn't. And at a minimum you're certainly free to use the modified total welfare standard if you want to. You call it like you see it, and we'll tell you if you've gone wrong".

My inclination is towards a total welfare test, even if it leads you down the odd uncomfortable path: the example everyone mentions is the Commerce Commission's 2015 wool scouring authorisation. There, there were consumer detriments (higher prices as a result of the market power of the merged-to-monopoly incumbent), offset by efficiencies from rationalising production facilities. The total welfare standard says money may have moved from consumers' to producers' pockets, but so what: the country's better off not tying up unnecessary resources in wool scouring plants. The consumer welfare standard says, get outta here.

Perhaps nobody will follow up the Court of Appeal's musings on the meaning of it all. But it is odd, to say the least, that this late in the game - the Commerce Act dates back to 1986, and its current purpose statement to 2001 - people are still wondering what, if anything, the legislators meant to achieve. Did the new purpose statement - "to promote competition in markets for the long-term benefit of consumers within New Zealand" - mean to enshrine a consumer welfare standard? If it didn't explicitly, can or should a consumer welfare standard be read into it?

There's a huge debate underway overseas about whether competition and merger enforcement has gone soft, particularly in the US: mergers, the argument goes, have been let through that shouldn't have been, and the big guys have been allowed to throw their market power around to ill effect. If you haven't read Thomas Philippon's The Great Reversal: How America Gave Up on Free Markets, now would be a good time. Against that background, maybe we should sort out our own thoughts in our own heads, too, and sit down and have a ground zero reassessment of what sort of competition law we ought to have.