It's a start

Last week's competition policy announcements by Scott Simpson, the Minister for Commerce and Consumer Affairs, were some modest steps in the right direction (and a nice little coup for the Competition Law and Policy Institute of New Zealand's annual workshop, where he made the speech).

He was surely right to make it easier for businesses to collaborate where there are likely to be net consumer benefits, given that the Commerce Commission's current authorisation process is "too complex, costly, and slow for business". Authorisation as it is currently practised isn't so much like using a sledgehammer to crack the proverbial nut: it's more like commissioning a squadron of war elephants to trample on it for months.

He also said that "We have also heard in your submissions" - the ones in response to MBIE's turn of the year consultation on 'Seeking feedback to improve competition in New Zealand' - "that businesses and individuals are increasingly reluctant to share information with the Commission because of fears confidential information could be released under the Official Information Act, potentially leading to retaliation or misuse of confidential information by competitors. This is undermining the Commission’s ability to collect evidence and receive useful information, particularly in investigations and merger clearances".

This is something that has been rumbling for a while, and he's done something about it. I can't say that I warm to one of his solutions, to exempt information provided to the Commission from the ambit of the OIA for 10 years, as I reckon we could do with less whiteanting of the Act,  but I suppose it will do the job in its way, and the other proposals (more flexible confidentiality orders, and whistleblower protection) look good ideas.

If I can be grumpy for a moment, I'd like to observe that this concern for confidentiality wasn't so obviously to the fore the last time Parliament looked at the Commerce Act, when in 2021 it introduced s99AA to allow the Commission to share information with "a public service agency, a statutory entity, the Reserve Bank of New Zealand, or the New Zealand Police", contrary to my elegant and well reasoned submission.

What struck me otherwise about the Minister's announcements was how much was still to come. The MBIE consultation was open-ended, but included a questionnaire (which many of us filled in) indicating what they most wanted to hear about. They included "the effectiveness of the current merger regime"; how to deal with "creeping acquisitions" and entrenchment of market power, for example, through 'killer' acquisitions of nascent competitors; whether to align our approach to mergers with the Aussies'; and a big question asking "How effective do you consider the current merger regime is in balancing the risk of not enough versus too much intervention in markets?". MBIE wanted to know whether potentially non-competitive mergers that were not notified to the Commission are an issue, and whether the Commission might need extra powers to deal with them (such as a stay or hold separate power). And MBIE wanted to know whether the Commission should be able to accept behavioural undertakings as part of a merger approval.

Outside mergers, MBIE also wanted to gather views about potential anti-competitive 'concerted practices' and what to do about them, and what role industry codes or rules might play in bolstering the competition regime. And then there were various rats and mice, including whether the Commission's injunction powers need modernising, views on assorted technical and minor amendments, and an open-ended "what else have you got to say".

So it would be fair to say that - six months after consultation closed - the announcements last week dealt to only two of the 11 broad topic areas MBIE canvassed: we've had some baubles, but no Christmas tree. The Minister said that "Further decisions on the merger regime, potential new industry codes, and other changes will be announced over the coming weeks". Given that we've been on the slow side in recent years when it comes to competition policy reform, compared with the Aussies in particular, a quick policy timeframe is good to hear.

How's my driving?

At the end of February the Commerce Commission published an interesting "look back" review of what it could learn from how its merger decisions have been made. As the Commission said, "The purpose of our ex-post merger reviews is not to determine whether the original decision of the Commission was correct or incorrect, compared to alternative decisions available to the Commission at the time. Instead, our reviews evaluate key factors of a decision and assess whether the Commission’s predictions have eventuated as expected".

That's fair enough. It's always going to be hard to figure out, with the passage of time, what would have happened (the 'counterfactual') if the Commission had made a different call. That's not to say it's a completely hopeless exercise. It could well be worthwhile at least looking for some (however inconclusive) indications of correctness: post-merger, did prices go up for reasons that weren't obviously related to the likes of input costs? Did innovation slow down? Did consumers see the benefit of the efficiencies the merger claimed to enable? But I accept that you may not get any clearly definitive answers, unambiguously linking the merger to unwelcome outcomes, and in that case the more limited objective of seeing whether you read the winds right at the time may be the best you can do.

And it is well worth doing. For one thing, there's been something of a global pushback against competition authorities that may have been too "soft" on mergers - wrongly seeing, for example, adequate competitive constraint on the post-merger entity when, in fact, there wasn't a realistic prospect of new entry or of expansion by existing competitors. For another, we don't do enough in New Zealand to go back and see if policies and plans and decisions worked out like we thought they would, and if not, why not. And finally it shows a pleasing openness to shed some light on the Commission's processes in an environment when some other agencies have been unnecessarily secretive (eg in their recent round of Briefings to Incoming Ministers) on what the issues are in their bailiwick and what they think about them.

The Commission has, apparently, been doing these post-merger reviews quietly in the background for a while: "The Commission first undertook some ex-post reviews of past merger decisions in 2015 [I'll come back to that one in a moment], focussing on cases between 2011 and 2013. A similar exercise was undertaken in 2016, looking at merger decisions made in 2013 and 2014. The Commission renewed the practice in 2019 with a review of six merger decisions made between 2014 and 2016. This process was replicated in early 2023, looking at merger decisions made between 2017 and 2019". But this is the first time the Commission has officially published its findings, and it covers the six reviewed in 2019 and the seven reviewed in 2023 (although in the end the Commission wasn't able to finish reviews of two of them, so 11 made it through to the final analysis).

The big takeaways? The Commission thinks that market participants are too blasé about the likelihood of new entry/expansion and about the ability and incentive, post merger, of consumers being able to exercise countervailing power, and it's going to be asking for harder (and preferably documentary) evidence on both fronts. And in dynamic markets - where there may be rapidly changing consumer fads and fashions (like tastes in the yoghurt market, in one of the reviewed cases) or where new technologies are being rolled out every other day - it may not make much sense to spend a lot of time on market definition, and it would make more sense to ask, post merger, will the merged entity be better able to get away with bad stuff or will it still face adequate competitive constraint.

The Commission said that this is the first time it has published the findings of these post-merger reviews, and strictly speaking that's true, but the results of the 2015 review are also in the public domain: I wrote about them at the time ('More on entry'). They were given as a presentation at the New Zealand Association of Economists' 2015 annual conference, very likely on the basis of "these are the views of the presenters not the Commission's". Interestingly, it came to similar conclusions about being suitably hardnosed about the likelihood of new entry, particularly where there may be high sunk costs (which might deter a potential entrant if entry were at risk of going pear shaped) or where entry is from overseas (particularly if they have bigger fish to fry than the New Zealand market).

Unfortunately the presentation isn't on the NZAE website (only a short and not very informative abstract), and while it used to be on the Commission's website, it doesn't appear to be now. If you're trying to track it down - I found it as an e-resource on the Auckland Library website, or you may have academic access to the paywalled economics journals - then you're looking for Lilla Csorgo and Harshal Chitale, "Targeted ex post evaluations in a data-poor world", in the 'Special Issue on Advances in Competition Policy and Regulation', New Zealand Economic Papers, 2017, Vol. 51, No. 2, pp136–147. 

Don't forget the benefits

Scrutiny of mergers is on the increase, notably in the US, where new draft merger guidelines have been widely interpreted as a sign of a more activist competition regulator proposing to take a tougher line, but also elsewhere. In the UK, for example, one economist recently wrote a piece in the Financial Times pointing out that the CMA's merger decline rate has been rising in recent years ('The UK’s competition watchdog risks undermining business dynamism', possibly $). 

In this latest swing of the pendulum, it's getting harder to argue for the 'good' merger, where the merged entity produces efficiencies (typically cost savings) or other benefits, such as innovative synergies from a combo that's more than the sum of its parts, or creates a more effective competitor to an entrenched incumbent. And more critics are finding more reasons to ping supposedly 'bad' mergers which reduce competition and increase corporate market power.

By happenstance, along have come two pieces of work, reminding us not to forget the 'good merger' story.

First, two hat tips for unearthing the first piece. One goes to the always interesting 'Blog Watch' column which the University of Canterbury's Paul Walker (aka GrumpyMcGrumpyface on Twitter) writes for the New Zealand Association of Economists' Asymmetric Information newsletter (if you're the proverbial intelligent lay person who'd like some very well-written takes on local and international economic issues, sign up for Asymmetric Information here, it's free). And the other goes to the equally approachable Conversable Economist blog, run by Timothy Taylor. He's also the editor of the Journal of Economic Perspectives, which "aims to bridge the gap between the general interest business and financial press and standard academic journals of economics" and is a terrific explainer in plain (or plainish) English of current economic debates (it's also free to read online, start here). Taylor's 'Recommendations for Further Reading' in each issue are always worth a look.

In his latest (July) 'Blog Watch', Walker picked up on one of Taylor's blog posts in April, 'After that Big Merger, What Happened?'. Taylor had come across some research done by folks at the International Center for Law and Economics, 'Doomsday Mergers: A Retrospective Study of False Alarms'. They looked back at six high profile, highly contested US mergers: their bottom line was that "Our retrospective analysis shows that many of the alarmist predictions of the past were completely untethered from prevailing market realities, as well as far removed from the outcomes that emerged after the mergers". With only one, partial, exception, the mergers had actually been 'good' mergers, with pro-competitive pro-consumer effects, or as Taylor summarised it, the retrospective case studies "do show pretty clearly that dire predictions about ill effects of mergers need to be taken with a few spoonfuls of salt" (he also wondered whether the merger sponsors' claimed benefits were as oversold as the merger critics' claimed costs were, which is fair enough).

The only partial exception was a big merger in the beer industry, where post-merger prices for some of the mass-market beers did increase (average prices remained steady). But that had the happy outcome that it created a profitable opening for the craft brewers, who have taken increased market share. And if you'd had the choice between a now more expensive but decidedly pedestrian beer and a tastily hopped artisan American Pale Ale, you'd have switched, too.

The other piece of research, which I came across on the ProMarket blog, is some work done for the World Bank. The ProMarket write-up is 'Firm Consolidations Hurt Workers, But Likely Not Because of Market Power', and the original all the bells and whistles World Bank working paper is 'Firm Consolidation and Labor Market Outcomes', very short summary here and full pdf here.

The researchers were primarily concerned about the adverse employment consequences of mergers, and they were well placed to investigate them. They were able to use a big administrative database in the Netherlands which contained matched employer-employee data, so they were able to compare what happened to employees in acquired companies after some 1,000 takeovers in the Netherland over 2011-15, compared to what happened at very similar companies that weren't taken over.

It's true that takeovers led to job losses: "There is substantial job loss among the workers of target firms: in the four years after a takeover, workers at a target firm are 8.5% less likely to be retained at the consolidated firm compared to workers in the control firm. This lower retention rate translates into income loss ... These effects are long-lasting and are present even in the fourth year after the takeover" (pp1-2 of the World Bank paper). 

Now, the researchers are, properly, concerned about this long-term adverse impact on those hit by involuntary job loss: my two best answers (which I've championed here before) are, at a macroeconomic level, maintaining as hot a labour market as you can run without triggering inflation, and, at a microeconomic level, 'active' labour market policies that make it easy to retrain, upskill, or go self-employed. Stomping on anti-competitive constraints in the labour market, like non-compete clauses, wouldn't go amiss, either.

But that said, the employment restructurings they are bemoaning are what in the competition policy game we would call efficiencies: they're cost savings, and as the researchers discovered, cost savings of a very specific kind. They found that if lab technician Kath in the acquired company is paid more than lab tech Rita in the acquiring company, Kath tends to get laid off. They also found that if there are lots of accountants in the acquired company, but the acquiring company already has lots of accountants, too, then the acquired accountants tend to get laid off. 

It makes complete sense, and it's likely to be a ubiquitous feature of mergers everywhere, not just in the Netherlands: if you were running the acquirer, you'd very likely act on similar lines. It's hard on Kath, and hard on the accountants, but the merged entity ends up more productive. And that's without thinking of any other efficiencies that may be on the table. Sure, for competition policy purposes, it's the net outcome that matters, and in any given case efficiencies may well be outweighed by detriments, but it would be silly to start from a viewpoint that efficiencies are nebulous or unlikely. A better starting presumption is that there are very likely to be at least some.

The Dutch example also got me thinking about what the New Zealand data might show. The Netherlands may well have a good administrative database, but so does New Zealand: indeed I'd hazard a guess that ours is top tier by international standards. Formally, it's the 'Integrated Data Infrastructure', or IDI: Stats calls it "a large research database. It holds de-identified microdata about people and households. The data is about life events, like education, income, benefits, migration, justice, and health. It comes from government agencies, Stats NZ surveys, and non-government organisations (NGOs). The data is linked together, or integrated, to form the IDI. The IDI complements the Longitudinal Business Database (LBD), which holds linked microdata about businesses. The two databases are linked through tax data".

At one point, access to the IDI was overtightly corralled, and not enough was being done to exploit its potential value. Now, it's being increasingly mined to good purpose: this year's NZ Association of Economists' conference featured a variety of IDI-based projects. AUT's Gael Pacheco and her team, for example, were able to follow (anonymised) Kiwi students who had done badly on the international PISA tests of numeracy and literacy to see what their subsequent employment, health, and justice system outcomes had been (as you'd expect, not good).

So why hasn't someone had a look at the effects of takeovers? It's all very well for supporters of mergers to claim benefits, and critics to claim costs, and the Commerce Commission to appeal to first principles of economics, but wouldn't we get more informed decisions if the question was, how does this proposed merger line up with what we empirically know about New Zealand mergers in general?

First guest post - Ben Hamlin on s47 penalties

I'm really pleased to announce the blog's first guest post - from Ben Hamlin, Barrister (Ben@hamlin.law). Our shiny new Commerce Amendment Act 2022 has got most attention for its (welcome) rewrite of s36 along Australian lines, but as Ben explains, there are other provisions which you need to be on top of. Ben's written about the increased penalties which will apply from May 5 to parties who breach s47 (anti-competitive mergers or acquisitions). As it happens, the Commerce Commission has been more active in recent years in opening s47 inquiries: Ben reviews the history and draws the lessons. Enjoy!

***

Increasing penalties for anti-competitive mergers

Merger activity is booming, so practitioners in that field might be forgiven if they had overlooked the forthcoming increase in penalties for mergers that contravene section 47 of the Commerce Act.

The increase is worth noting, but also provides an opportunity to reflect on merger enforcement in recent years.

The potential penalties for mergers are on the up

The first substantive provisions of the Commerce Amendment Act 2022 come into force on 5 May 2022, including an increase in the corporate financial penalties associated with mergers that substantially lessen competition. From 5 May onward, the maximum penalty will be the greater of:

(i)                  $10 million:

(ii)                 either,—

(A) if it can be readily ascertained and if the court is satisfied that the contravention occurred in the course of producing a commercial gain, 3 times the value of any commercial gain resulting from the contravention; or

(B) if the commercial gain cannot readily be ascertained, 10% of the turnover of the person and all its interconnected bodies corporate (if any) in each accounting period in which the contravention occurred.

As is often the way, the change in merger penalties barely rated a mention in the passage of the Commerce Amendment Act 2022. At the third reading Hon Andrew Little, speaking in the place of Hon David Clark, noted:[i]

Another important way that the Act protects the competitive process is through its prohibition against anti-competitive mergers or acquisitions. For this prohibition to be effective, it needs to be able to deter entities that may benefit significantly, in commercial terms, from the merger. The bill ensures this can happen by increasing the monetary penalty the courts can impose if entities are found to have contravened the prohibition.

This change brings the penalty into line with contraventions of Part 2 of the Act, and broadly into line with the penalties for a breach of the equivalent provision in Australia.[ii] The increase from $5,000,000 to $10,000,000 is also broadly in line with inflation since the penalty was set in 1990.[iii]   

So far, very orthodox: deterrence through ensuring that the potential benefits are outweighed by the potential penalties.

Does an increase in non-notified merger investigations suggests inadequate deterrence?

An interesting further justification was provided in the Cabinet paper approving the decision to increase the penalty. The Courts have cautioned against placing much weight on cabinet papers when interpreting legislation.[iv] Any judges reading may wish to avert their eyes.

The paper noted that: [v]

“Since the beginning of 2018, the Commerce Commission has undertaken investigations into eight possibly anti-competitive mergers for which clearance or authorisation was not sought. This suggests that the current penalties for anticompetitive mergers are not acting as a sufficient deterrent.”

Does it? As the readers of this blog will be aware, the penalty is only part of the optimal deterrence equation. The probability of detection and prosecution is equally, if not more, important.

The Commission’s merger enforcement 2010/11 to present

The Commission’s published merger statistics show that there was indeed a real spike in s47 investigations.[vi] In the seven years between 2010/11 and 2016/17 there were a total of 11 section 47 investigations decided. Yet in the two years 2017/2018 and 2018/2019 a further 11 section 47 investigations were decided.

Financial year	10/11	11/12	12/13		13/14			14/15			15/16			16/17		17/18		18/19	19/20		20/21		21/22 HY	Total
Clearances (s 66)
Total applications decided	10	10	8		12			14			12			7		9		11	9		8		10	120
Unconditional clearance	8	8	6		9			11			9			3		5		8	9		6		6	88
Clearance with divestment	2	1	0		2			1			2			0		1		2	0		2		1	14
Declined	0	1	1		0			2			1			3		2		0	0		0		0	10
Withdrawn	0	0	1		1			0			0			1		1		1	0		0		3	8
Section 47 investigations
Number decided	1	3	2		1			2			1			1		7		4	1		2		0	25

Table 1 Source: Commerce Commission Merger Determination and Enforcement Statistics - December 2021

What could have caused the jump? It is difficult to distinguish correlation and causation. But many competition law tragics will remember a heady period where the Commission blocked five transactions in 13 months: Sky/Vodafone February in 2017, Aon/FPIS in March 2017, the NZME/Fairfax Authorisation was declined in May 2017, Vero/Tower in July 2017, and Trade Me/Limelight in March 2018. Appeals were filed against three of the decisions (Sky/Vodafone, NZME/Fairfax, Vero/Tower), but two were withdrawn and the NZME/Fairfax appeal proceeded unsuccessfully.

It is possible that blocking a number of high profile mergers in quick succession caused some parties to avoid clearance. In any event, the increase in section 47 investigations did not escape the Commission’s attention. In August 2018, the Commission announced its priorities for 2018/2019, and ‘non-notified mergers’ was among them. The Commission’s then Chair was quoted as saying:[vii]

“New Zealand is one of a few jurisdictions with a voluntary merger clearance regime and the Commission is seeing an increase in non-notified mergers. Over the past 2 years we have opened five investigations into non-notified mergers. The success of a voluntary regime relies on the credible threat of enforcement proceedings so we will act quickly in these cases to prevent adverse impacts on competition in markets.”

Opening investigations is, of course, the easiest part of the process, and does not mean any competition issue exists. Competition investigations can start, stop, or spend a long time in between, and are often a black box to the outside world. Up until July 2017, little information appears to have been published unless a section 47 investigation resulted in litigation.

Fortunately, in July 2017, the Commission had announced that it intended publish a record of section 47 investigations on its website, to ensure the public and market were aware of investigations into potentially anti-competitive transactions.[viii] As a result, the Commission’s case register records outcomes for 11 section 47 investigations since the beginning of 2018.

A breakdown of them is instructive:

• In three cases, the Commission considered there was either no issue, or insufficient evidence to establish a competition issue: Bondor NZ/The Long Group, Bejier Ref AB/Heatcraft NZ/Kirby NZ, and Glenninburg Holdings/Wallace Group GP.
• In four cases, the Commission closed its investigation following a divestment or change in transaction. In Vero Insurance NZ/Tower Insurance, no further action was taken after Vero divested the 19.9% of Tower it had bought during its attempted takeover of Tower. In Fulton Hogan/Stevenson, no further action was taken after Fulton Hogan agreed not to acquire Stevenson’s Huntly quarry. In David Ferrier/Cavalier Wool Holdings no further action was taken after David Ferrier sold down his interest NZ Wool Dumping. In Datix/RL Solutions, no further action after Datix divested several customer contracts.
• In four cases, the Commission commenced proceedings. In Platinum Equity/OfficeMax Holdings, a settlement was reached after litigation was commenced. Platinum purchased OfficeMax but divested its WINC NZ business. In Wilson Parking NZ/Penrith Holdings, a settlement was reached after litigation was commenced. Wilson Parking offered an enforceable undertaking to divest the assets it had acquired. In First Gas/Gas Net, an agreed civil pecuniary penalty was imposed. And most recently, in Objective Corporation/Master Business Systems, a civil pecuniary penalty is to be imposed following settlement.

Without going into the merits of any of these individual cases, the outcomes overall suggest that the Commission was right to have some concerns but appears to have been able to address them. While headlines tend to focus on declines and penalty cases, divestures can equally represent a significant outcome for competition. In some cases, such as Platinum Equity/OfficeMax litigation, or the recent Ampol/Z Energy clearance, the entire existing New Zealand business is divested. It is also clear that some deals are ultimately stopped because of competition concerns, which can see investigations halted or clearance applications withdrawn.[ix]

This burst of section 47 activity may not be permanent. The Commission’s register indicates that while three section 47 investigations were commenced in mid-2020, there have been none commenced since August 2020 despite a veritable boom in merger activity. Instead, the 2021/2022 year appears to be one for the books in terms of merger clearances, with 15 applications decided in the first three quarters.

One possible explanation for this is that the Commission’s announced focus on non-notified mergers, followed up with investigations, has resulted in the pendulum swinging back towards parties seeking clearance. The increased probability of detection may have been what was missing, rather than the size of the penalties.

Conclusions

One busy year is not enough to draw firm conclusions, and we are working with small numbers in any event. But I would suggest three conclusions can be drawn, for what they are worth, from the data we have.

First, the Commission’s decision to regularly publish data on its merger work, and the decision to make section 47 information public on its register, assists in analysis and understanding of the Commission’s work in context. 10 years ago, a practitioner would need to rely on insider knowledge, intuition, or a lengthy series of OIA requests, to attempt to understand what was going on.

Second, the Minister was right that there was an increase in section 47 investigations, but it does not necessarily follow that a penalty increase was needed for that reason alone. If in the future there is a decrease in voluntary notifications, perhaps because of the cluster of blocked transactions, then the appropriate response might be more investigation. That may require resourcing and appropriate enforcement tools, rather than another increase in penalties.

Third, the Commission’s performance should be assessed in context. The Commission has not declined a clearance since Trade Me/Limelight, more than four years ago. But that does not mean that the Commission has not been busy enforcing New Zealand’s merger laws in other ways. Anyone viewing New Zealand as a soft touch could be in for a surprise.

And from 5 May, it is potentially a much larger one.

Ben Hamlin, Barrister.

Disclosure: Ben Hamlin was Deputy General Counsel, Competition, later Chief Legal Adviser, Competition, at the Commerce Commission between February 2017 and March 2022. His views are his own.

---

[i] Commerce Amendment Bill, Third Reading, 17 March 2022.

[ii] See section 76 of the Competition and Consumer Act 2010 (Cth).

[iii]  The RBNZ Inflation Calculator indicates that a basket of goods worth $5 in Q2 1990, when the Commerce Amendment Act 1990 was passed, would be worth $9.60 in Q4 2021, the most recent available data.

[iv] See for example, A Labour Inspector v Southern Taxis Ltd [2021] NZCA 705 at [51].

[v]  Cabinet Paper “Review of Section 36 of the Commerce Act and Other Matters: Policy Decisions” (18 February 2020) at [56], available online here.

[vi] Merger determinations and enforcement statistics – December 2021, available here.

[vii] Commerce Commission “Commission releases 2018/19 priorities” (press release, 9 August 2018) available online here.

[viii] Dr Mark Berry, “Opening remarks” (Competition Matters 2017, Wellington, 21 July 2017).

[ix] For example, the recent Cargotec/Konecranes and Aon/Willis Towers Watson mergers were both cancelled because of competition concerns.