Rebuilding a Better AIG
The new CEO's background inspires confidence that he can fix issues.
American International Group (AIG) has been a troubled company since the financial crisis, although its worst problems then primarily related to its financial products division, and all the issues that led to its blowup and subsequent government support having been reduced to a point of immateriality.
Still, AIG would have destroyed substantial shareholder value even if it had never written a single credit default swap, as noncore businesses needed to be shed and even its core operations had serious problems.
Some investors understandably have grown frustrated with the still-poor overall performance, but considering the variety and depths of the company’s issues, we don’t view it as a surprise that AIG’s turnaround is a work in progress. In the meantime, the market continues to offer AIG at a deep discount to book value, and while we expect a turnaround to take time, modest improvements in returns should help lift the stock higher.
Peter Hancock, who was CEO from October 2014 through May 2017, initiated a number of actions to improve performance and did have success on some fronts. We think the most notable improvements were divestitures of noncore operations, realization of cost efficiencies, and aggressive capital returns. The overarching goal of these actions was shrinking the business to a more manageable and profitable base. But he failed to solve AIG’s biggest issue, which is the poor underwriting performance of its commercial property and casualty lines, and this ultimately led to his departure.
Hancock led a push to materially lower costs and improve efficiency, which included some difficult decisions, such as reducing senior management positions by 20%. To some extent, because the business has been in retrenchment mode, some cost reductions were necessary just to stay level. However, even on a percentage of revenue basis, the company has made significant strides in improving its cost structure. General operating expenses (excluding one-time items) as a percentage of operating revenue declined to 17.4% in the first nine months of 2017, compared with 19.4% in the first nine months of 2015. We estimate that this improved annualized return on equity by 130 basis points, if we exclude deferred tax assets from the calculation.
Hancock also oversaw a significant number of divestitures. Given the overdiversification of the business and the depth of its operational issues, we believe an aggressive push to simplify the business as much as possible was in order, and the correct divestiture targets were chosen.
AIG was effectively a ward of the state in the years following the financial crisis, and it built up excess capital during this period. Significant divestitures and reduced levels of activity in core businesses freed up further capital. As a result, Hancock aggressively returned capital to shareholders, primarily in the form of stock repurchases. Given that the stock traded at a material discount to book value during the past few years, we saw these repurchases as potentially further leveraging the operational improvements the company could make in the coming years. To put this plan into perspective, from the start of 2015 through the present, AIG’s total capital return equated to just over half of the company’s market capitalization at the start of 2015.
While we think Hancock deserves some credit for the progress that the company made under his watch, arguably the improvements that he effected--reducing costs, divestitures, and returning capital--were the low-hanging fruit in AIG’s turnaround efforts.
Hancock essentially failed to make any significant progress in tackling AIG’s biggest problem--its record of poor underwriting in commercial P&C lines--as he was unable to put an end to the company’s history of adverse reserve development. This is understandable, as Hancock did not have a background in underwriting; he joined AIG in 2010 following a career in banking. We largely agree with the overall strategy Hancock initiated to turn the business around, but results suggest his lack of an underwriting background was a major obstacle in implementing it successfully. As such, we did not view his departure as a step back for AIG, but potentially a step toward realizing a true turnaround.
Commercial P&C Key to Acceptable Returns
While AIG remains a highly diversified insurer, we see its commercial operations as the key to achieving acceptable returns. The commercial P&C segment has the largest equity base and therefore is the biggest driver of overall returns.
Due largely to the cost efficiency improvements AIG has made, returns in the company’s life and consumer P&C insurance operations have already moved to what we think is an acceptable level, and we believe that the scope for error in these businesses is relatively limited. While the commoditized nature of the life and consumer lines limits opportunities for sustained excess returns, underwriting is a relatively straightforward process, which reduces the potential for major problems. The company’s runoff operations do eat up a significant portion of capital, and this is likely to be a slight drag to overall returns. But returns have not been particularly poor, and equity attributed to runoff business has fallen to $10 billion from $19 billion since the start of 2015.
We think the improved results across the company’s other segments make it clear that the only major obstacle to AIG achieving acceptable overall returns is addressing the poor profitability in its commercial P&C operations. With that in mind, we think it is critical to understand the nature of the problems that AIG has endured in this area.
AIG Has Been a Terrible Underwriter
AIG’s P&C underwriting results over the past decade have been basically the worst in the industry. While its issues have been concentrated in commercial P&C lines, the poor returns there have overwhelmed overall results. Not only has AIG posted one of the worst average combined ratios over the past 10 years, its results have also been particularly volatile. Even worse, the company has consistently generated significant underwriting losses in a period when industry pricing and underwriting results have generally improved, largely in response to low interest rates and reduced investment income. AIG has not shown any meaningful positive trend over this period. The primary reason for both the poor average level and the volatility has been consistent and at times large adverse development charges.
AIG has been dramatically out of line with industry trends when it comes to reserve development. Most P&C insurers have consistently recorded favorable development over the past decade because of low inflation, whereas AIG has consistently posted adverse development and taken a number of large reserve development hits. We estimate that adverse development over the past five years has negatively affected AIG’s combined ratio by about 7 percentage points. If AIG had actually posted favorable reserve development in line with its peers, its combined ratio would have been roughly in line with industry averages. As such, we believe the key to bringing AIG’s commercial P&C results to acceptable level is to eliminate adverse development.
We believe reserve development is the key metric in assessing underwriting prowess, which in turn is the key consideration for our stewardship view. We see reserve development (at least on a relative basis) as primarily a stewardship issue.
In our view, AIG’s history on this front showcases the two potential causes of underwriting underperformance. First, management can show a lack of discipline, prioritizing growth over profitability. In the short term, this approach can look good, as it takes time for the overly aggressive underwriting assumptions used to justify that growth to become apparent. But shareholders are ill-served in the long run. The second way a company can fall short is in terms of underwriting acumen--that is, it can simply be relatively bad at assessing claims risk. This leads the company to take a disproportionate amount of unprofitable business, as its more skilled competitors take advantage. In our view, while AIG has improved its underwriting discipline in recent years, its acumen remains lacking, and this is the reason the company’s reserve development issues have persisted.
In the years leading up to the financial crisis, AIG made a big push into long-tail casualty lines (the "tail" of an insurance line refers to the length of time it takes for claims to be ultimately settled). This decision resulted in more than a decade of reserve development issues. AIG saw strong growth but also took on asbestos and environmental liabilities, which proved costly over time, and underestimated the claims it would ultimately pay in riskier areas such as excess workers’ compensation. In our view, this was a classic case of poor underwriting discipline, as the company’s prioritization of top-line growth led to aggressive assumptions and long-term profitability issues. Still, AIG has since actively moved away from the most troubled areas, and it is only a matter of time before the company starts to outrun these issues.
AIG has dramatically reduced its underwriting volume in recent years, with the pullback concentrated in the areas that have been problematic. Net written premiums for commercial lines are down about 30% from 2014 levels. As a result, it would be difficult to say that AIG continues to overemphasize growth or that discipline is still an issue at this point.
However, recent reserve development suggests that adverse development has been shifting toward newer business, and moving past the company’s historical development issues may not simply be a matter of time. The components of reserve development over the past three years have continually shifted toward business written after the company began to shrink. Before 2015, most of AIG’s adverse development related to policies written before the crisis. But in 2015, more recent accident years made up a significant portion. In 2016, more recent policies made up the majority. In the third quarter of 2017, recent policies made up almost the entire charge.
We think this trend suggests that AIG’s issue at this point is its underwriting acumen, with the company failing to properly assess the risks it has taken on in recent years. To some extent, AIG’s dramatic efforts to shrink the business may have exacerbated this, as the company appears to have disproportionately retained less attractive business (probably because it was the easiest to retain).
Regardless of the exact cause, the persistence of adverse development shows that the company still has serious work to do in addressing its underwriting acumen.
AIG’s Issues Aren’t Structural
In our view, while economic moats are possible in some lines, insurance is at its heart a commodity industry in which it’s difficult to achieve sustainable excess returns. AIG’s historical performance could be taken as a sign that there are areas that are particularly difficult to operate in, and if so, management’s ability to improve returns going forward could be very limited. However, delving into the details and looking at peers operating in similar business lines shows that AIG’s commercial P&C issues aren’t structural, and that this area should actually be the company’s strength, not its Achilles’ heel. This provides further support for the idea that the root of AIG’s issues is mismanagement.
Compared with most other commercial P&C operators, AIG’s business is characterized by its size, international footprint, and focus on serving large corporate clients, with narrow-moat Chubb (CB) (formerly ACE Limited) being the company’s most comparable peer. When we break out AIG’s commercial P&C operations by line, the mix of the two companies is remarkably similar. Given that Chubb has historically produced some of the best underwriting results in the industry, it is hard for us to argue that AIG’s mix of business is the problem. Even looking at narrow-moat Travelers (TRV) and W.R. Berkley (WRB), which concentrate more on middle-market clients, the mix of business is not that dissimilar, mostly driven by the different policy needs of middle-market clients.
Looking closely at underwriting results by business line, we see that AIG has seriously underperformed its peers on a line-by-line basis. AIG has badly lagged the industry in general liability, workers’ compensation, and commercial auto--lines that collectively make up over half of its book. Chubb, on the other hand, has generally outperformed the industry in the same lines. This suggests that a properly managed AIG should be more than capable of producing acceptable results.
If we look at AIG’s commercial P&C operations from a qualitative perspective, the company does look to have the makings of a moat in this area, and Chubb’s success there provides a model to follow. In our view, the primary way for a P&C insurer to give itself a competitive edge is to focus on the least commodified areas of the insurance market. The term “specialty line” is often used in the P&C insurance industry to describe this type of line. While the term is used loosely and is not precisely defined, some areas such as excess and surplus lines are clearly specialty. AIG is actually the largest writer of excess and surplus business in the U.S., although it has reduced its participation in recent years. In our view, specialty lines can be something of a double-edge sword, in that they are fundamentally attractive but can be dangerous for relatively unskilled participants.
But other areas qualify as specialty, in our view. We would classify specialty lines as being any line with unique risks that requires extensive experience or unique assets to underwrite effectively. Both AIG and Chubb have historically focused on serving large, global clients. Often, these companies have risks that do not fit nicely into one country or one particular policy line. To service complex accounts, an insurer must be able to match these unique and diverse risks with an equally diversified global network and have the balance sheet to take on large individual risks. Only a handful of insurance companies have the size and footprint to compete with Chubb and AIG in this area.
It's difficult for us to find a reason beyond mismanagement to explain AIG’s poor historical performance in commercial P&C lines, and the record at Chubb demonstrates that a well-managed entity with a footprint similar to AIG’s can be very successful. So, in our view, the solution to AIG’s most pressing problems looks quite simple: Find the right manager to improve underwriting acumen.
Duperreault Has Already Built the Company AIG Wants to Be
Brian Duperreault took over as AIG’s CEO in May 2017. He started his career at the company, serving in various roles at AIG and its affiliates over 20 years, during its heyday. He left AIG in 1994, serving as CEO (and eventually chairman) of ACE Limited (now Chubb) for the next decade. Duperreault came out of retirement to run the struggling insurance broker Marsh & McLennan (MMC) in 2008, leading that company over the next four years. In late 2013, he came out of retirement again to run Hamilton, a small Bermuda-based insurer, before being tempted away to AIG.
We see Duperreault’s record at ACE as the most relevant portion of his background. When he took it over, the company was a small operator with only about $400 million in annual premiums. From this starting position, Duperreault was arguably the primary architect in building the franchise into the premier operator that the company is today. Further, the company’s business mix by the end of his tenure was roughly in line with AIG’s today in terms of lines, geographic footprint, and client base. As such, we think the record he established at ACE looks portable to AIG.
It’s difficult to find fault with the results Duperreault achieved at ACE, as his tenure was marked by strong growth and superior underwriting profitability, and the company’s ongoing strength following his departure suggests he left a very solid foundation. He held the chairman and CEO role from 1994 to 2004, and then stayed on as executive chairman until 2006.
Over this period, ACE underwent a dramatic transformation. Driven in part by some large acquisitions, the company’s premiums grew at a 32% compound annual rate and increased by a factor of 37. From a very modest starting position, Duperreault built out a leading commercial P&C position that brought ACE into head-to-head competition with AIG.
More important, amid this growth Duperreault built a franchise that generated superior underwriting profitability, firmly establishing the company as both a large and moaty business. After 1994, ACE produced a lower combined ratio than the industry every year and on average outperformed the industry by 9 percentage points.
In our view, Duperreault’s record shows he is more than capable of running a large commercial P&C operation successfully. The key difference is that in taking over AIG he now needs to focus on improving an existing operation as opposed to building one from scratch. His experience with Marsh & McLennan demonstrates that he is up to the task.
Duperreault came out of retirement in 2008 to take the CEO position at Marsh & McLennan, a leading insurance broker. Marsh & McLennan was struggling at the time, in the wake of the controversy over contingent commissions that led to criminal charges against some Marsh executives in 2004. From 2003 to 2008, operating margins had declined from 22% to 7% (excluding a goodwill impairment charge). By the time Duperreault departed in 2012, operating margins had recovered to 15%. While insurance brokerage is structurally a very different business, we think Duperreault’s proven ability to step into a troubled situation and improve results is relevant to his prospects at AIG.
Duperreault’s Plan Is Simply to Improve Underwriting
We don’t expect a grand transformational plan for AIG, but rather a common-sense approach to restructuring the company’s businesses, relying on Duperreault’s experience to bring them more in line with standard practices at more successful peers. We like that he seems to be focused squarely on AIG’s main issue--its underwriting issues in commercial P&C lines. Based on Duperreault’s comments and our conversations with the company, we see the changes that he expects to effect as falling along some basic lines.
Retain less risk. Duperreault has been clear in his belief that AIG has been overly aggressive in retaining risk. While its size, business model, and large corporate focus necessitate a large risk appetite to some extent, he thinks it has been overdone, with a relatively large hit from natural catastrophes during the third quarter driving home the negatives of AIG’s past approach. We expect to see a greater use of reinsurance. While reinsurance is not value-creative in and of itself, we view this as a positive move, as reducing large losses will help to reduce volatility and avoid events that will sap confidence in the company’s turnaround efforts.
More active and effective use of data and analytics. Duperreault is known in the industry for his strong interest in using technology and data to drive better underwriting results. Given its size and the breadth of its operations, AIG should, if anything, have a data edge over most of its competitors. However, we think any potential edge AIG might have accumulated over the years has been squandered historically as data collection was not centralized. Hancock had led a push toward centralization during his time at AIG, but recent results suggest the improvement has not filtered through the organization. Given Duperreault’s strong interest in this area, we expect this to be a primary focus.
Fewer generalists, more specialists. Duperreault appears to believe that AIG’s size and diversification has allowed underwriters to drift into areas outside their circle of competence to the detriment of the company overall. He has established a unit that will handle all loss-sensitive business and has refocused underwriters at Lexington, AIG’s excess and surplus business, strictly on these lines.
P&L accountability. In the past, AIG followed something of a matrix approach in its organizational structure, which dispersed accountability across the organization. While the insurer had been adapting its organizational structure in recent years, Duperreault intends to give discrete underwriting units full control of and accountability for their profit and loss statements, in order to encourage profitable underwriting.
In our view, these actions would largely represent a move toward common practice at the higher-quality companies in the industry. AIG historically has had unusual corporate structures that clearly did not work well; at a minimum, we believe that a move toward a more industry-standard structure should help the company move toward industry-average results.
The Market Has Not Priced In Any Potential Improvement
Mismanagement has been the primary driver for AIG’s underperformance in commercial P&C lines. Duperreault’s successful record in this regard, and his relatively common-sense proposals so far, give us reason to believe it is realistic to assume some modest improvement going forward, especially considering AIG’s starting point.
The market, however, has not reacted favorably to Duperreault’s appointment. Since AIG announced he would be joining as CEO, its shares have fallen 2%, compared with a 16% gain for the S&P 500. Among domestic insurers with market capitalizations of $10 billion or higher, AIG trades at the lowest price/tangible book value by a good margin, and its multiple is roughly half that of the average insurer. Its current valuation equates to 0.8 times tangible book value, a level we believe is consistent with expectations of an extended period of value destruction. As such, we believe the market is not pricing in a material probability of significant improvement under Duperreault, despite his record.
We think two related factors are driving investor disappointment. First, AIG was very aggressive about returning capital to shareholders under Hancock, providing precise goals around capital returns. Duperreault has moved away from offering precise targets and has noted that capital return is not necessarily his key priority. Second, Duperreault has said his intention is to grow AIG, and to this end, he is open to strategic acquisitions.
We can understand the market’s skepticism on this front, but we believe investors may have misinterpreted some of his comments. AIG was always going to pivot to growth at some point, but the company’s size and the maturity of the industry were always going to be limiting factors, in our view. We don’t expect to see AIG realize strong growth, but merely stop aggressively shrinking the business. Further, we believe that Duperreault’s growth comments might be directed more for internal consumption, to raise morale among employees who have endured years of reductions and layoffs.
At first glance, acquisitions would seem to be premature, given the deep problems in AIG’s existing operations. However, based on our conversations with the company, any acquisitions would be focused on bolt-on deals in areas that are complementary to businesses already performing at reasonable levels, such as life insurance and consumer P&C. We think a valid argument can be made for supporting these segments with strategic acquisitions.
It was inevitable that capital returns would slow at some point, as AIG worked down its excess capital. We do believe the stock is materially undervalued, which raises the bar for acquisitions to be a better use of capital, but we are skeptical that Duperreault will pursue any major deals. We expect that he will return most of free cash flow to shareholders, but that he wants some flexibility on this score, which we believe is reasonable.
Turnaround Will Take Time, but Some Factors Limit Near-Term Risk
It will take time for better underwriting to work its way into reported results, and given AIG’s history of adverse development, we believe there could be a lag between the time the company stops recording significant adverse development and the market starts getting comfortable with the idea that AIG’s reserves are sufficient.
On the positive side, the scope for major adverse development hits in the near term looks limited. Historically, the company has concentrated its formal review of reserve levels in the fourth quarter, which has made the end of the year the riskiest period for investors. While the company has tried to distribute this process more evenly throughout the year, Duperreault has pushed a greater portion of the review process up into the third quarter. As a result, AIG has gone through a formal annual reserve review process on 80% of its book through the first nine months of 2017, which is partly responsible for the $836 million in adverse development reported in the third quarter. With most of this review process completed ahead of the final quarter of the year, a large charge in the fourth quarter is not impossible, but the potential is meaningfully diminished.
Additionally, following the large adverse development charge in the fourth quarter of 2016, management acted to mitigate its reserve exposure by signing a significant reinsurance deal with Berkshire Hathaway (BRK.A)/(BRK.B). Under the terms of this deal, Berkshire Hathaway will absorb 80% of any future adverse development on certain business. This deal does have important limits. It covers only policies written in 2015 and prior and covers only certain lines (primarily the long-tail lines that have plagued AIG in the past). Notably, the deal covered none of the $836 million adverse development recognized in the third quarter. In our view, this deal primarily provides protection against the lack of underwriting discipline at AIG before Hancock’s tenure, not the underwriting acumen issues we think AIG has more recently faced. Still, we think the deal materially lowers near-term risk, as it essentially mitigates a large portion of potential adverse development risk--it covers $34 billion in reserves, or almost half of AIG’s total reserves. This would be a poor deal for Berkshire Hathaway if AIG doesn’t move past its historical development, and we take some confidence from Berkshire’s implicit bet that the worst of AIG’s reserve development issues are behind it.
Given the current valuation, long-term prospects under new management, and the limitation on potential adverse development charges in the near term, we believe AIG’s current market price tilts toward reward for investors willing to be patient--acknowledging, of course, that the company is not without risk.
Brett Horn does not own (actual or beneficial) shares in any of the securities mentioned above. Find out about Morningstar’s editorial policies.