As yet, the European Commission (EC) has not produced a formal State of Objections to Aon's proposed takeover of Willis Towers Watson following the Phase II competition probe it began in December.
But sources are increasingly emphasizing the likelihood that the European competition authorities will require remedies to greenlight the transaction.
This building sense that the deal will face regulatory challenges that could either extend the timeline, or derail the transaction altogether can be tracked by the merger arb spread, which has widened over the last two months from around 6% to 10%.
It also seems clear that the parties are now at the critical stage of the process, with the clock stopped for just over a month to date for further information gathering, and a Statement of Objections due in around the next 10 days.
As such, it seems likely that Aon – as with any organization in this position with a strong track record around execution – will be war-gaming different scenarios as it seeks to figure out what the minimum threshold moves would be to successfully address the EC's objections.
Multiple market sources said they were not yet aware of any outbound activity from Aon or its bankers to explore possible divestitures, with such activity frequently coming after remedies are agreed, according to legal sources.
In the face of skepticism from industry sources initially centered on the combined market share in reinsurance, Aon has throughout maintained that it is well placed to close the acquisition of Willis without divestitures.
The line was repeated on Aon's Q4 earnings call. Reiteration of this position underscores the difficult balancing act Aon has around communication to different stakeholders. A confident line is likely to put Aon to support negotiations with regulators, while long-term investors – and possibly staff – may prefer to prepare for the possibility of divestitures.
Competition concerns around the P&C parts of the transaction, which will make Aon the largest broker in the world with revenues of around $20bn, seem to center on three areas:
Each of these areas have different characteristics, pointing to scope for differentiated outcomes. (Sources have highlighted some areas of concern in human capital and benefits, including investment advisory, but they are out of scope for this piece.)
Both the EC and by inference the New Zealand antitrust authorities have flagged large corporate clients as an area where the deal might reduce competition.
"The Commission's initial market investigation identified a number of concerns in relation to the supply of commercial brokerage services especially to large multi-national customers, who depend on brokers with a high level of expertise and a global presence," the EC said in a December 21 note when it initiated its stage two review.
The same concern was evidently raised by New Zealand’s Commerce Commission in a Statement of Issues document based upon the Aon-Willis response which has been made public in redacted form. In that note, the parties sought to address issues flagged with the implications for competition for "large customers and those with more complex mandates".
Sources suggested the stopping of the clock with the EC relates to the need to secure additional information from multi-national clients about the way they purchase insurance and the insurance broking options available to them.
The crucial question here is whether or not large account business is a market or, to pose it differently: do large multinational corporations have distinct needs that make them overwhelmingly select one of the Big Three?
First of all, it is clear the biggest corporations in the world do overwhelmingly choose to purchase insurance via the Big Three, although no public market share figures are available.
Senior broking sources have estimated their market share for this client base at 70%-80%, although they stress that Marsh and Aon are the dominant players, with Willis a distant third.
"Willis is much more of a mid-market broker," one senior source noted. "But they get invited to the RFPs."
It is not clear though that market share should be assessed based on a stratification of clients by size and complexity – i.e. that this client base has truly distinct needs – and Aon/Willis pushed back heavily on the idea that large accounts constitute a market.
Aon and Willis said in the New Zealand submission: "The Commission does not attempt to articulate what constitutes a ‘large and/or complex commercial insurance customer’. This may be reflective of the evidence that shows there is no clear, or meaningful, distinction between ‘large and/or complex commercial insurance customers’ and other customers.”
It added later that "the needs of relatively large clients are not sufficiently distinct from those of other clients to justify defining a separate market and, regardless of client size, the same competitors can and do compete to fulfil those needs".
The parties push back on the idea that there are a significant number of multi-national clients buying a single global programme. Instead, they stress the decentralization of risk management, with jurisdictions and risk lines broken up in the purchasing process, meaning they "procure insurance using a combination of multiple broking firms (in addition to insurance channels, i.e. the direct channel)”.
In addition, it argues that the risk classes purchased by a large client "are no different from those sought by other clients". Moreover, the parties point to a range of carrier support which could help a broker to service a multi-national client.
There is some force in these arguments, but it remains the case that the fees spent by the largest multinational clients go to Marsh and Aon, and to a much lesser degree Willis, which must raise questions around the service propositions of other brokers for these customers.
The question about whether competition regulators judge large account business to be a marketplace is particularly crucial – not due to the revenues attaching to the business (although they would be meaningful) but because Willis is not structured in this way.
Willis is organized by product line and geography. As such, if the EC decides large accounts are a marketplace and competition is harmed here, it is difficult to see how Willis could divest.
Theoretically, internal assets could be reorganized, a leadership team appointed and staff segregated within different offices for allocation to a large accounts team – a process that would involve staff being removed from smaller accounts that they worked on. After this process, the large accounts business could be sold to the likes of Gallagher or Lockton.
But the degree of reengineering would be significant and highly complex – with the people management side likely to be particularly difficult – and at minimum the H1 deal timetable would be heavily impacted.
Reinsurance has also been cited by sources as an area of regulatory focus, and publicly stated by regulatory authorities including those in Australia. The deal brings together the number one reinsurance broker, with the number three reinsurance broker and it does so two years after the number four was itself taken out by two.
This looks like a piece of “three to two” consolidation given that the drop to the current number four and five takes you to businesses that are just a fraction of the size of the number one and two players.
Sources estimated the brokered marketplace at around $5bn of revenues, equating to a 90% market share for the Big Three if you include the facultative reinsurance revenues of Willis, which do not currently sit within Willis Re, but would presumably be merged in post-deal.
This breaks down as a ~36% market share for Aon, ~34% for Guy Carpenter and ~20% for Willis Re (including the fac revenues). There are no up-to-date figures available for TigerRisk which would be the new number three but its 2019 revenues were $125mn. The business has been rapidly growing but it is hard to see how its market share could exceed 4% (i.e. <$200mn).
As previously outlined by the Inside P&C research team, merging Aon's reinsurance broking business and Willis Re would create significant concentrations of brokered market share – well over 50% – in a range of international markets. These include major European markets like France and Germany, as well as Japan, Australia and some Latin American countries, according to sources.
The crucial counterpoint is that these are brokered market share figures, and there is no reason to treat the brokered market in isolation. The true market around risk transfer for cedants also includes the direct channel and alternative capital.
These points are, naturally, stressed by the parties in their New Zealand submission. In addition to scope for cedants to turn to the direct markets or ILS, it further stresses that cedants can discipline brokers by threatening to switch to another broker, or to retain additional risk.
"Importantly, a cedant's decision is not binary and cedants can effectively discipline brokers simply by threatening to allocate some portion of their risk to other channels," Aon and Willis explained.
Sources have stressed that, for the most part, brokers are remunerated in reinsurance based on standard levels of brokerage, and even if there are brokerage sharing or rebate deals, price competition is not what typically drives decision-making. Instead, considerations include the capabilities, expertise and market relationships of the broking teams vying for the business.
The other aspect in the decision-making is, however, problematic from a competition perspective.
Multiple senior sources at a combination of cedants, reinsurers, reinsurance brokers and retail brokers have over many years described the important role played by the trading relationships on the inwards book in determining the allocation of shares broking the outwards protections.
Market sources said that by increasing the size of Aon's retail business and creating two retail giants, the reliance on just two players for origination grows, making it more difficult to eschew Aon and Guy Carpenter in favor of other reinsurance brokers when allocating the most important shares on their outwards programmes. This would be the case regardless of the talent, capabilities and market relationships that a TigerRisk or a Gallagher Re were able to bring to the table.
Sources stressed that discussions of "tying" are not held with brokers and that no agreements – formal or informal – exist on the subject. However the perception remains in the C-suites of cedants that inward flows are influenced by decisions on which intermediaries are chosen to handle the outwards.
Regulators seem to have turned a blind eye to the issue indefinitely, or else they are simply comfortable with the way the industry currently operates. But this feature of industry structure will become more pronounced if Aon acquires Willis’s reinsurance broking arm, which at least creates scope for the issue to be considered afresh.
If regulators are concerned about reinsurance broking competition, there is a range of possible remedies which the parties could offer to try to satisfy them, with a number of possible counterparties.
1. Willis Re-plus could be sold – This would include $725mn of revenues from the Willis Re business, an additional ~$300mn of fac which sits in Willis' insurance business, and potentially revenues attaching to technology/services businesses sold to cedants that sits in the same segment as Willis Re. The options to dispose of this would run to: a) sale to a single competitor i.e. Gallagher, Lockton or a PE-backed US retailer like Hub, USI, Alliant, AssuredPartners, NFP etc, b) a listing of the company, c) sale to a private equity house or consortium.
2. Willis Re could be sold – Willis Re as it is currently defined (the $725mn treaty business) could be disposed of in its entirety. This would bring the same acquirers into play, although it would make it an easier deal for the smaller names.
3. Willis Re International could be sold – Willis Re is organised into three segments: international, North America and specialty. Sources believe the more pronounced market-share issues are in international markets like Europe, Japan and Australia. There would potentially be complications, particularly around global accounts that could be serviced out of offices that span the divide – but it might be achievable, particularly given the relatively small headcount that would need to be divided. Such a deal would require a sale to a competitor with existing infrastructure and market presence.
4. Specific portfolios like France, Germany or Japan could be sold – Such a deal or deals would require sales to competitors with the existing infrastructure and broader market presence to allow them to viably compete. Most in the market question whether this could meet a threshold for maintaining competition, but reinsurance broking does include star players and teams that are able to move business. There will be a real-world test of this in the coming weeks as Willis bonuses are paid on a stagger on each side of the Atlantic, potentially paving the way for defections to the likes of TigerRisk, Lockton Re and Gallagher Re.
There is a great deal of speculation around what would be required to satisfy the concerns that the EC may have on reinsurance, but there is no clarity at this point.
Aon will want to hold onto as much of the revenues and talent as possible, and will further want to deny it to competitors, allowing you to build something like a spectrum of best-case to worst-case scenario (assuming it has to do at least something).
Its worst-case scenario is likely to be a sale of Willis Re-plus to AJ Gallagher, the fourth-biggest broker in the world. Gallagher has long seemed to be the most natural home for Willis Re in an enforced divestiture scenario, and the impression that is preparing its guns has been strengthened by its move yesterday to file shelf registration documents that would allow it to rapidly issue equity, debt or prefs.
Although this potentially provides Gallagher with a once-in-a-lifetime chance to secure a leading reinsurance broking franchise, this is a $3bn-$4bn deal (even at a discount multiple) and well beyond anything it has done in its career for scale. It would also need to find a way to get comfortable with the risk that the Willis Re management team – a number of whom personally control significant revenue – would commit to their new owner.
If they are judged "suitable acquirers", other brokers could create competitive tension, but what is a big deal for Gallagher's is a transformative transaction for Lockton, and other top 15 US retailers backed by PE including Hub, USI, Alliant, Acrisure, AssuredPartners and NFP.
Floating Willis Re or selling it to a private equity house would be a less damaging scenario for Aon, as it would not strengthen a competitor that Aon competes with in other areas.
However, disposals via listings are rare and a traditional IPO in New York would be expensive for a business the size of Willis Re, and a SPAC deal potentially unacceptable to the regulators.
Private equity has a track record of buying forced disposal assets, according to regulatory lawyers, although they stress that the European competition authorities have cooled on them in recent years owing to the perception that they will not invest for the long term. PE could look to allay concern around their intentions by providing a commitment to hold the business for 10 years, but this would shrink the pool of potential sponsors.
PE likes the broking sector, but Willis Re is not the most obvious asset for a sponsor given that it is a high-quality business that lacks obvious M&A opportunities, denying it an obvious PE playbook.
Partial or portfolio sales would be the best outcome for Aon if remedies are required given that they would allow the acquirer to hold onto a significant amount of the business, including in areas where Aon currently has lower market share.
The EC in its statement of preliminary concerns identified a number of specialty lines where it was concerned that the deal could reduce competition.
It named "financial and professional services", credit and political risk, and cyber and marine – specifying that this was for large multi-national customers.
And it separately pointed to brokerage for all customers in space and aerospace manufacturing, "as well as in a few additional risk classes in specific national markets".
Willis has global product lines in areas like marine, aerospace and finpro. They are not separately constituted businesses – but they are units with their own leadership, staff and a segmented client base.
Concerns that regulators may have in these areas parallel the EC's effective insistence that Marsh sell JLT Aerospace when it acquired that business in 2019.
Parallel transactions could be carried out in one or multiple areas in this case, with the staff and assets dropped into a competitor with the necessary infrastructure.
There would be a range of potential acquirers for Willis specialty units with their center of gravity in London including Howden, Gallagher, Lockton, Ardonagh, Corant, BMS, Miller and Acrisure – creating scope for significant competitive tension that would help valuations.
The crucial point here would be which companies the EC judged to be a "suitable acquirer". When MMC was obliged to sell JLT Aerospace, it is understood that only Gallagher and Lockton were judged to meet that mark – something which is probably inconsistent with the market's perception about how big a parent would have had to be to maintain competition.
Amputation
As long suggested, forced disposals are the biggest threat to the value creation inherent in the deal, which absent those looks huge owing to a mix of major strategic and financial benefits for Aon.
Significant disposals would make the deal less compelling both by curbing the strategic and financial benefits, and also by extending the period of messiness for staff and investors (and to a degree clients) by delaying the close.
Aon's management team has committed itself wholeheartedly to the deal, staking its reputation on the transaction. And indeed, throughout, their story-telling around the transaction has emphasized not the value of the individual pieces or the scale of the transaction, but its scope to allow the broker to innovate and build new capabilities to seek to address the industry’s waning relevance to clients.
And the base-case disposals cap in the merger agreement has been set high at $1.8bn (19% of Willis revenues), again suggesting real commitment on the acquirer side to getting the deal through even in the face of regulatory turbulence.
Amputating a limb to save the body may be an acceptable choice.
Inside P&C provides unparalleled market intelligence on the entire US P&C market – from small commercial and personal lines right through to reinsurance and Bermuda. Redeem your complimentary 14-day trial for more premium content from Inside P&C.
Scan here to download the app