Description

AON is an asymmetric way to invest in one of the few areas of the global economy that looks likely to improve in 2012:  Insurance.

Summary Investment Thesis

1. After almost seven years of declining pricing, the P&C insurance cycle appears to have troughed and may be turning up. 
2. At ~10x current year EPS it is cheap absent an improvement in the cycle.
3. Investing in a broker doesn’t involve the balance sheet complexity of analyzing a carrier, where the assets are bonds paying low yields and the liabilities are reserves against premium written at a time of weak pricing.
4. Brokers, unlike carriers, will benefit from price increases even if they are only sufficient to offset declining investment income caused by long-term ZIRP.
5. There are some company specific initiatives, in particular GRIP, that suggest scale may be more valuable over time.
6. Management is very good and will use free cash flow intelligently, including returning excess capital to shareholders.
7. The stock did nothing in 2011, despite incrementally positive news in the industry.

 Business overview

AON is the world’s largest insurance broker, and a leading human resource consulting firm.  In the brokerage segment, the company matches commercial customers with primary insurance providers in addition to matching primary insurers with reinsurance providers.  Given the consolidation in the brokerage space over the past two decades, the industry is becoming more oligopolistic with only Aon, Willis, and Marsh able to fulfill the needs of large global companies.  The human resources segment involves benefits consulting, retirement services, talent strategies, and HR outsourcing.  The largest competitors here are Mercer and Towers Watson.

Insurance brokerage and human resources consulting are attractive businesses with low capital requirements, strong barriers to entry, and sticky client relationships.  Insurance brokers add value by matching fragmented buyers of insurance with fragmented suppliers, helping commercial customers design their insurance needs, and enabling insurance providers to bid for business on an apples/apples basis.  Brokers are either paid a commission by the insurance provider or a fee by the commercial customer (typically larger companies pay fees and smaller companies buy insurance gross of commissions).  Perhaps the best evidence of how sticky clients in the brokerage business are is that when Marsh ran into trouble after the Spitzer investigation, the company lost a tiny amount of business even in the face of aggressive poaching efforts by Willis and AON.

From a secular perspective, the principal fear we have is potential fee compression for the brokerage space over the long term.  We have a similar fear for commercial real estate agents, residential real estate agents and virtually all professions that charge as a percentage of transaction volume, but this concern seems to have been alleviated over the past decade as fee compression in the insurance brokerage space has flattened out.  This is likely due to consolidation in the brokerage space, compared to a still fragmented insurance underwriting industry.  However, the weak cycle over the past several years has presumably also supported revenue capture, as carriers bid for business and commercial clients on a fee basis don’t cut fees even as premium pricing declines.  For what it is worth, AON’s response to this fear is, “Of course there could be fee compression, that is why we have to provide two dollars of value for every one dollar clients pay us.” In any event, this is a long term issue that is not likely to manifest itself until/unless we are in a much better market.

HR consulting/outsourcing encompasses a mix of businesses, but the general theme is assisting corporate customers with ongoing HR and compliance requirements, as well as project work, which is often driven by regulatory change.  The business has good margins and decent growth and AON likes being in the “risk and human capital business.”  There is some synergy between the two in that the person you deal with at a corporation about sourcing insurance is often the same person or group that hires human resources consultants.  Consulting should have low but steady growth, and given Greg Case’s background at McKinsey he should be able to do well with it.

In mid 2010 AON announced that it was acquiring Hewitt Associates (“HEW”), one of the largest HR consulting firms.  While the deal was only modestly accretive to GAAP EPS, it was 15% accretive on an economic basis (i.e. excluding amortization of acquisition related intangibles).  Strategically, the jury is still out.  AON probably had to do something, as its existing consulting business was subscale in a number of areas.  However, the most recent quarter was a sloppy one for the segment and it remains to be seen how the two function together once they are fully integrated.

1.      Waiting for Godot…is the cycle finally turning?

We discussed some of our thinking on insurance cycles a year ago in the WRB write up and Q&A.  It’s probably worth a quick look there.  Broadly speaking, we think insurance follows a vaguely consistent cyclical pattern.  Most of the time pricing is in modest decline, carriers talk themselves into bad risks, employees chase business to keep their jobs, and underwriters extrapolate unsustainable positive trends into the future instead of assuming mean reversion.  Then, there is some sort of catalyst, the problem is recognized, and there is panic.  Pricing moves up sharply as carriers must not only improve accident year profitability but also repent for past sins.  It is impossible to know for sure if the cycle is turning, but we think there are a growing number of sign posts:

The strain:

• It is clear that pricing is inadequate on an accident year basis in most lines of insurance.
• Reserve releases are ending.
• 2011 was likely the worst year in history for global catastrophes.
• Lower and lower investment yields are hammering profitability.

The scuttlebutt:

• If you listen to commentary from Bill Berkley or Traveler’s you would think pricing is rising fairly significantly.  They are talking about prices being up mid to high single digits, and have varying degrees of confidence that this will accelerate in 2012.
• AON is less optimistic.  Their data indicates the pricing is declining by less on a year over year basis.  This trend got better every quarter last year.
• These data points are not necessarily inconsistent.  When a cut-rate competitor “buys business” and takes a piece away from WR Berkley, it shows up negatively in AON’s data but not in WRB’s.
• Marketscout, a leading data provider, reported a 1% increase in industry wide U.S. P&C rates in November; the first y/y increase since February 2005.
• Lots of industry managements we speak with seem incrementally positive (although starting at varying levels of optimism/pessimism).
• But…January reinsurance renewal pricing was just okay.

The conclusion:

• By almost any cut, it is clear pricing has stopped coming down and is at least flattening.
• It is possible that in a world where bond yields are so low that insurers should earn lower ROEs because there is a lower cost of capital.
• I feel comfortable saying pricing is unlikely to decline further and there is a reasonable chance it gets better.

2.      Valuation & return profile

Because AON earns money as a percentage of global insurance premiums written, the depressed economy and insurance cycle have pressured revenues significantly.  Pricing has been in a soft cycle for a number of years and the amount of exposure companies have has declined as well.  This has led to AON exhibiting poor organic growth over the past few years but this now appears to be inflecting.  Although the brokerage industry grows at about worldwide GDP over time, it is likely that the next several years could be above trend because the growth is lumpy and is based on the insurance cycle. 

From an upside perspective, revenues inflecting up while margins go from 20% to 25% in the face of the company buying in stock will likely be powerful.  The company is only moderately leveraged, has bought in $5BB of stock over the past six years, and has a CEO running it who owns over 1mm shares of restricted stock and 1mm options and is looking to maximize shareholder value.  As the insurance cycle flattens out and starts to improve, margins expand, and stock buybacks continue, a company like AON could get re-rated from 10x EPS to 13-15x EPS while earnings rise from $4.50/share to $6/share a few years out.  This would be roughly a double in the stock.

The downside should be reasonably well protected because the stock is already cheap and hitting 2012 numbers only requires modest top line growth due to the impact of ongoing share buybacks and ~$90MM of incremental synergies related to the Hewitt transaction that are rolling through on a 2012 versus 2011 basis.  The two principal downside risks to this are continued softness in the Consulting business and fx translation.  As a reference point, the stock bottomed in the $35 zip code during late 2008/early 2009.  Since then, the company has generated a lot of cash although a direct comparison is tricky because of the Hewitt transaction.

3&4.    Insurance brokers versus carriers

We have investments in both carriers and brokers.  However, AON is our largest because it seems like the most asymmetric. 

• Analyzing the balance sheets of carriers is hard.  Although we like some of the property and casualty insurers, AON is less risky because it does not have a bond portfolio that could get marked down with higher interest rates, has no potential reserve problems, and will benefit from inflation while insurers would be hurt by it.
• From a timing standpoint, it has lagged most of the carriers we like during 2011 and so it is arguably cheaper/embedding a low probability of a turn.
• As the Fed pursues long term ZIRP, this is putting enormous earnings pressure on financial institutions that generate earnings by leveraging fixed income assets.  P&C insurers fall into this category.  For a typical P&C insurer, with 3-4:1 leverage, a 200bp change in the return available from fixed income assets translates to a 5-6% change in ROE (on-shore insurers usually have some of their investment portfolios in muni bonds and so effective tax rates are well below 35%).  Prices need to rise high single digits just for insurers to recover the earnings they are losing from this dynamic.  AON should benefit from this, while insurers are just digging out of a hole.
• That said, in the slingshot hard-cycle case, the best positioned carriers are likely to outperform AON as the operating leverage in the model is much higher.
• We prefer AON to WSH and MMC mainly because we think it is cheaper and better run. 
  • WSH is a pure play, but trades at a large premium on an EV basis and is de-levering.  As for management, let’s just say Joe Plumeri is controversial.
  • MMC has a similar business mix to AON, but trades at a premium and has less cyclical upside in brokerage because more of its business is with very large customers and is fee based.

5.      AON is starting to leverage its scale

AON places $40B of retail commercial P&C premiums globally, so it has more information flow than anyone in the business.  Until recently, the company had never made an attempt to capitalize on its information advantage, but over the past few years it has developed the best technology in the industry in the form of their Global Risk Insight Platform (GRIP).  This is a compelling innovation in that it gives AON’s brokers the best tools and information in the business.  The company’s brokers are now able to tell customers real-time the cost of various insurance policies, current market conditions, the lead underwriters for the region, and best practices from competitor companies in a whiz-bang form without the client having to wait a few days or take the broker’s word.  For example, a broker can walk into your office now with a laptop and say “here is $500mm professional liability policy; here are the three main carriers; two are in NY, one in UK; here are pricing terms and conditions over last 45 days for companies similar to yours; have you thought about directors and officers insurance or workers comp in addition to the professional liability because the 100 most similar companies to you take it; and, here are the average levels of exposure for all similar companies to you around the world.” 

This is a potentially dramatic change to an industry that has been very much about handshakes and relationships, where brokers have all their contacts and information in their head and pitch to customers with “I know where the markets are at and I know a guy at x insurance firm that can help you out.”  Now, on a daily basis, 7,000 Aon account executives are entering every possible piece of information about insurance bids into AON’s GRIP system.  You may not need $40B in premiums to have this product but clearly there is a meaningful amount of scale necessary to have enough data to make the results relevant and to invest in the technology.  Moreover, it takes a cultural change in an organization to get all the brokers to give up their “secret sauce” to the corporation for the greater good, thus limiting the individual broker’s power over the longer term.  Interestingly, management is also using this to attack the fee leakage in AON’s business.  Whereas on average AON is supposed to get 10% commission rates, the brokers are not charging the right amount a lot of the time and AON wants to recapture that lost commission.

6.      Management & capital allocation

Greg Case joined in 2005 from McKinsey, where he was the head of its financial services practice.  When he joined, AON was an amalgam of over 450 acquisitions that had not been integrated.  Case immediately initiated a restructuring program, paring off its underwriting businesses and rationalizing its cost structure.  Over the past five years, he has increased margins in the brokerage segment from 15% to 20%, more than doubled EPS, divested low return businesses, and acquired and integrated Benfield, a top international reinsurance franchise.  Additionally, management did this in the face of a declining insurance cycle, a nasty recession, and declining interest rates reducing its float profitability.  The company also repurchased almost $5 billion of stock (although less on a net basis).  Shortly after closing the HEW transaction, the company began making progress towards repurchasing the 64 million shares issued to HEW shareholders.  While the company needed to issue shares to maintain its investment grade credit rating, we believe most of its free cash flow can be used to repurchase shares as the balance sheet is only moderately levered at 1.6x net debt / EBITDA.

We have been impressed by the culture of AON versus other companies in the industry.  When we ask them a question they give a thoughtful response.  They seem to evaluate capital allocation decisions from a return on invested capital standpoint and look at everything compared with buying in AON stock.  When asked about acquisitions, it sounds like there is little in the U.S. that will work from an ROIC standpoint vs. buying in stock but there are smaller international franchises that they may be able to leverage worldwide.

7.      AON has been a laggard

While there is always a story, the fact is that AON stock has lagged recently.

Price change from 12/31/10

Versus the other global brokers….

AON    +0.5%

MMC   +12.6%

WSH    +11.4%

…or high quality P&C firms.

AON    +0.5%

WRB    +26.8%

CB       +16.7%

MKL    +7.7%

Appendix: The pension liability

As of 12/31/10, AON’s combined U.S. and international pension plans were underfunded by approximately $1.66BB pretax, or $1.2BB after-tax (we believe after-tax is the more correct way to analyze because the principal balance can be paid down with pre-tax dollars unlike actual debt.)  This figure is relatively static absent interest rate or asset price changes, as the vast majority of the plans are closed.  The underfunding would have risen in 2011 given lower rates and poor asset returns, but we estimate company contributions of almost $500 million should offset this.

One should arguably add the underfunding to the enterprise value when doing EV/EBITDA valuations and somehow deal with when looking at P/E multiples.  However, $109MM of pension expense already ran through the P&L in 2010.  This is 9% of the $1.2BB “after-tax” figure, so arguably there is already a 9% interest expense being expensed through earnings.  Because of this we don’t think current earnings are overstated, although the pension fund will be a short term cash headwind again in 2012 as contributions will be larger than the GAAP expense.

Catalyst

• Insurance cycle turns up
• Large share buybacks
• A few clean quarters as the HEW deal integration approaches the finish line
• Possible multiple expansion for differentiating initiatives like GRIP