The years since the financial crisis have shown that American International Group AIG would have destroyed substantial value even if it had never written a single credit default swap, had noncore businesses it needed to shed, or had material issues in its core operations that it needed to address.
We believe underwriting discipline is the key stewardship factor in the insurance industry, and precrisis management's focus on growth is the root cause of AIG's poor historical performance. The shift to focus on risk-adjusted returns and operational efficiency set a course in the opposite direction. We think the strong underwriting background of current CEO Brian Duperreault positions him to make progress on the one big issue AIG still needs to resolve. We don't see any structural issues with AIG's business, and we think the franchise can earn adequate returns under capable management. The process will take time, but Duperreault promised to return AIG to underwriting profitability in 2019 and is on track to achieve this goal.
AIG's historical results have been plagued by adverse development from certain long-tail commercial casualty lines. The company has posted adverse reserve development every year since 2008, while its peers have generally been posting favorable development due to low inflation. At the end of 2016, AIG had to boost reserves by $5.6 billion, a move that we believe led to former CEO Peter Hancock's departure. In the past couple of years, adverse development has moved into recent accident years, suggesting the underwriting culture at AIG needs a fundamental fix.
While AIG's history highlights how reserve development problems can be annoyingly persistent, and the company took another $800 million slug of adverse development in the third quarter of 2017, we believe AIG can move past this issue, given Duperreault's record of building a successful underwriting culture at AIG's closest peer. Further, AIG’s recent reinsurance deal with Berkshire Hathaway helps to mitigate this risk going forward, as Berkshire will take 80% of reserve development charges in troubled lines.
Areas of Strength, but Not Enough for a Moat In general, insurers do not benefit from favorable competitive positions. Industry competition is fierce, and the products are essentially commodities. Furthermore, most participants do not know their cost of goods sold for a number of years, allowing them to underprice policies without knowing it. Firms have a significant incentive to chase growth without regard for profitability, a cycle that repeats itself as competitors are forced to match artificially low prices or risk losing business.
We do not believe AIG has a moat, and while we see considerable scope for improvement in its results over time, we think it would be very difficult for the company to develop a moat in its current form. AIG does have some areas of strength: It is among the largest commercial property and casualty underwriters, which means that it should be able to leverage its proprietary database to better price and select risks than smaller peers, and recent investments in technology should allow it to better exploit this potential advantage. AIG also has one of the widest geographic reaches, which gives it an advantage in garnering business from global firms. But it participates in essentially every major insurance line, across every distribution channel, and in a number of geographies. As such, we believe that any advantage the company might hold in particular areas is likely to be diluted elsewhere. Furthermore, we don't believe there are any meaningful synergies to be realized from housing P&C and life operations under the same roof, although the combination will remain necessary for some time in order to make full use of AIG's tax-loss carryforwards. We think AIG's diversification will make it difficult to consistently beat industry returns in the long term.
Balance Sheet Is Sound The singular risks that AIG faced during the financial crisis are now largely in the past. Like other insurers, AIG's biggest risks are claims in excess of the amount reserved or material impairments in its investment portfolio. The company continues to report adverse reserve development from past underwriting decisions in long-tail lines, and it is unclear when it will move past this issue. AIG also has a material exposure to catastrophe losses, which are inherently unpredictable. Within its investment portfolio, its largest exposure is to corporate debt, and it could suffer from losses in this area if the economy suffered another recession.
We think AIG's balance sheet is sound, although the company is arguably in a somewhat weaker position than peers until it can improve profitability. Equity/assets was 14% at the end of 2018. This level is lower than P&C peers but reasonable if AIG's life insurance operations, which operate with higher balance sheet leverage, are considered. During 2014, the company reduced its debt load by about $10 billion and eliminated much of its high-coupon debt, which improved its financial flexibility. Barring any unforeseen events, we think the company has room to continue to return capital to shareholders.
AIG returned $11.8 billion, $12.8 billion, and $7.4 billion in dividends and repurchases in 2015, 2016, and 2017, respectively. Duperreault has signaled that returning capital is no longer the top priority, and previous buyback levels were unsustainable, but we believe the maturity of the business might necessitate healthy levels of capital return going forward.