As the menu of investment options has expanded over recent decades in an increasingly complex market, insurance companies have watched the investments they make with their policyholders’ premiums grow in complexity, too. The insurers’ tastes have evolved from traditional securities to ever more tantalizing—and risky—financial instruments, including mortgages, private equity, junk bonds, and collateralized debt obligations. During the bull market, that approach seemed more than reasonable; the pursuit of ever-higher yields was apparently a surefire way to boost returns. Yet as the portfolios and risks of the insurers grew, few of them updated their systems and management processes to oversee the potential downside of the new investment strategies. With more than half of the insurers’ risk capital supporting investment activities by the time the bull market peaked, in the late 1990s, many of these companies resembled little more than hedge funds with insurance businesses on the side. But the insurers’ performance metrics, such as net income and return on equity (ROE), were the same ones they had relied on in the 1960s—and inadequate for the complexity of the risk they were bearing.¹

No surprise, then, that when a brutal bear market combined with a groundswell of corporate-bond defaults² US insurers faced billions of dollars in losses from defaults on collateralized debt obligations and troubled bond issuers. Ratings agencies downgraded 151 and 148 US property and casualty insurers in 2002 and 2001, respectively, compared with 77 and 59 in 2000 and 1999, respectively, and a similar trend affects life insurers.³ European insurers endured similar pressures, losing billions of euros from their general accounts—in some cases as a result of exposure to stock markets, in others because of asset-liability mismatches.

Many insurance strategists now find themselves flummoxed as they search for a more stable business model to ensure steadier and more reliable returns. Clearly, greater operational effectiveness and better distribution of insurance products and underwriting of insurance risk will be at the center of any broad insurance industry revival. In addition, our work⁴ indicates that it is critical for the industry to rethink its exposure to investment risk if it is to avoid further calamities. By crafting an investment strategy that eschews the pursuit of higher yields and focuses instead on generating clear value, insurers can reduce their exposure to unnecessary risks and, in the process, free up billions of dollars in capital.

For most insurers, such a strategic reorientation will require a fundamental shift in the way they invest. First, insurance product groups and the investment function must have separate responsibility for their returns, costs, and risks. Second, to maximize the value from investments and to free up risk capital, companies must refocus their investment strategies to better match the risks of their assets with those of their liabilities rather than try to generate high yields. Finally, insurers should ground their investment professionals more solidly in actuarial science, risk management, and portfolio management. These changes will make the performance of the underwriting and investment functions more transparent and thus highlight which products are creating value and which are destroying it.

A flawed investment strategy

By the rather staid standards of the insurance industry, investment strategies in recent years have been a walk on the wild side

By the staid standards of the insurance industry, recent investment strategies have been a walk on the wild side. North American life insurers, for example, reduced the quality of their bond portfolios; their holdings in bonds rated BBB+ and lower made up 39 percent of the $2 trillion under their management in 2001, up from 23 percent in 1996. European insurers also took more risks, but the way these companies are regulated meant that they concentrated their exposure in equities rather than corporate bonds. Indeed, by 2001 some large European insurers had committed as much as 50 percent of their assets to equities, compared with 4 percent and 22 percent for US life insurers and property and casualty insurers, respectively.

This shift has been gradual and in some companies, perhaps, not even intentional. Chasing ever-higher yields seemingly provided easy money, and as long as the economy was booming the markets appeared to reward insurers that increased their risk levels. But when equities slumped and bond defaults rose, many insurers were left with considerable losses.

Moreover, the long bull market also obscured the fact that many insurers’ investment strategies yield limited shareholder value and often destroy it. In the process, these insurers tie up substantial amounts of capital that could be put to use more productively for shareholders in the underwriting of value-creating insurance risk. Three approaches at the heart of many insurers’ investment strategies illustrate the problem.

Increasing the investment portfolio’s risk profile

When credit defaults were low and equity markets buoyant, insurers generated attractive short-term returns from riskier securities. But while increasing the riskiness of assets may return higher investment income over the life of individual securities, it rarely creates real value for shareholders once the cost of risk is factored in. Indeed, the higher yield of riskier securities rewards the security holder only for bearing higher risk and price volatility—a trade-off that shareholders could execute on their own.

Further, regulations require companies to hold more risk capital for riskier investments—a cost that amounts to approximately $2 billion a year for the US life insurance industry. That makes it difficult in practice for insurers to create shareholder value by increasing the risk level of their investments.

Mismatching assets and liabilities

Insurers also commonly purchase assets—typically bonds—that mature over a longer period than their liabilities, in effect betting that they will capture some of the higher yields of these securities. If interest rates behave exactly as the market expects, with assets and liabilities marked to market,⁵ insurers earn enough to meet their underwriting obligations and pocket the intended profit margin. If interest rates rise faster than expected, the market value of the bonds declines and the loss on their sale eats into profits. Only if interest rates rise more slowly than expected do insurers earn enough to meet their underwriting obligations and return additional earnings to shareholders.

Purely by chance, half of those who bet that interest rates won’t move in the way they are expected to will likely be right—but they are unlikely to be right consistently. In the US treasury market, the world’s most liquid and efficient financial market, betting against expected interest rates is like betting against the house at a casino. When one leading US insurer, faced with today’s low interest rates, had to purchase a multimillion-dollar hedge to protect itself against the risk of rising interest rates, it wiped out years of revenues gleaned from mismatching its assets and liabilities.

Trying to beat the market

Warren Buffett, the CEO of Berkshire Hathaway, is the rare investor who can claim to have consistently created value for shareholders by beating the market through stock picking. Insurers have assumed that by actively managing their own investment portfolios, they too can identify assets that are mispriced and offer higher returns without taking additional risk. Yet except for the less-liquid asset classes (for instance, private equity), the evidence belies that assumption (Exhibit 1). To make good on it, the insurers would need to have research, analytical, and trading capabilities equivalent to those of the best hedge funds and asset managers. In addition, they would have to deliver returns in excess of the market benchmark consistently while still operating under the demanding liquidity requirements and regulatory constraints specific to insurance portfolios. Active management also carries the costs of frequent trading, fees paid to external managers, and an additional tax burden from the realization of capital gains. These costs would chip away at whatever outperformance the insurers achieved.

Investing for value

Both to satisfy short-term expectations and to maximize the creation of long-term economic value, insurers should stop trying to generate short-term yields and instead return to a more disciplined approach—assessing investment strategies on a risk-adjusted basis and taking into account investment expenses and total capital costs. If insurers did so they would allocate their risk capital to investment activities more efficiently and free up risk capital for what they know best: pricing and underwriting insurance risk. An insurer can diversify away that risk by pooling it and hence hold it at a lower cost than an individual investor could. Thus, a renewed focus on product pricing and underwriting can offer considerably more opportunity to create excess returns for shareholders than trying to beat the much more transparent and efficient financial markets. To adopt this superior approach to investments, the insurers will need to do three things most of them haven’t done so far.

Create degrees of separation

First, product and investment groups must have separate responsibility and accountability for their costs, returns, and risks. On the investment side, this responsibility includes managing the optimal portfolio required to support insurance products and bearing the risks of any deviations from that optimal portfolio. Today, though the investment and underwriting sides of most insurers are managed separately, the two bear collective responsibility for the overall profitability and risk of products. The investment group gets credited with the overall performance of the portfolio rather than the risk-adjusted performance in excess of the optimal liability portfolio. Product groups bear both insurance and investment risk charges. Such collective responsibility obscures the underlying performance and risk inherent in insurance products. Both investment managers and product managers can thus take credit for strongly performing products and blame each other for poorly performing ones.

A precedent for a more transparent approach can be found outside the insurance industry: in the early 1990s, banks divided accountability for their balance sheet performance between their asset and their deposit-gathering functions. The industry also broadly adopted more sophisticated capital-management tools, such as risk-adjusted return on capital (RAROC) and shareholder value added (SVA), in large part because massive bankruptcies of banks around the world during the late 1980s led to a radical rethinking of the management of financial risk. The greater transparency and accountability notably improved the industry’s performance. A lot of banks began pricing their products more accurately and making better decisions about their product lineups; many smaller and midsize banks, for example, left the corporate-lending, mortgage, and consumer-credit-card businesses as a result of sustained underperformance. In the decade since, the US banking industry more than tripled its return on assets, improving its return on equity by 50 percent, and at the same time strengthened its capitalization to reflect risks more accurately.

But insurers have been slow to adopt such a structure, in part because the complexity of their products makes the development of sophisticated performance-management tools, such as SVA, more difficult.⁶ Moreover, until recently the industry hadn’t faced a challenge as dramatic as the one the banking industry faced in the late 1980s. Insurers have suffered by coming late to the game of sophisticated performance management: the banks’ performance exceeded that of the S&P 500 over the past decade, while insurers have disappointed the markets and lagged behind the index.

Yet several insurers have broken from the pack, dividing accountability for their balance sheet performance between product and investment groups. Products are designed on the basis of fully matched investment portfolios—in other words, on the assumption that investment managers will fully match the risk of underwriting and that they alone bear the consequences of additional investment risk, such as asset-liability-mismatch risk or additional credit risk. The resulting transparency allows management to be clearer about the sources of value in both underwriting and investments, to make better pricing decisions, and to base the allocation of capital on the potential of products to create economic value. Adopting this approach made one large North American insurer aware that though its fixed-annuity business generated a large share of its earnings, the business was destroying value for shareholders because of the enormous economic risk capital it required.

Greater transparency for the insurers’ underwriting and investing sides doesn’t mean that product groups and investment managers can operate in isolation; indeed, at the end of the day, they must fit together perfectly. The pricing of products, for example, is closely linked to the expected returns of the investment portfolio, while the investment strategy is closely tied to the structure and nature of the liabilities inherent in product design.

At insurance companies that have separated the responsibility and accountability for underwriting and investments, collaboration is intense, constructive, and seamless. In this environment, transparency can create a high level of trust, which makes it easier for management to reach difficult decisions, such as pricing products out of the market when the disciplined investment strategy behind them can’t be both competitive and profitable. Transparency also lets companies take advantage of changing conditions when rising prices make some insurance products more competitive.

Manage investments strategically

The second component of the new approach to investing is the development of a clearly defined strategy to help the investment function weigh risk against value creation. This strategy should include rigorous standards for what types of risk an insurer is willing to accept, how much exposure to each type is acceptable, and what return the company expects from taking these risks. In practice, such a strategy can be implemented through three different kinds of programs. Strategic asset allocation (SAA) defines what asset classes the insurer wants to invest in and how to allocate assets to each class. General account portfolio management (GAPM) specifies the investment strategy (the asset classes, the mix, the duration of securities, and their credit quality) for a product. Asset liability management (ALM) defines the amount of interest rate risk a company is willing to take and ensures, on an ongoing basis, that this limit is respected. Although most insurers already have such programs in place, their form and execution vary enormously. Asset allocation may be based on experience and rudimentary calculations at some insurers, for example, whereas others use leading-edge portfolio-optimization software and implement rigorous processes.

The interdependence of value, risk, returns, and capital requirements leads to a surprising conclusion: don’t attempt to maximize yields

Viewed from a long-term perspective, the interdependence of value, risk, returns, and capital requirements leads to a counterintuitive conclusion: insurers looking to maximize shareholder value should match their risk exposure from investments with the financial risk inherent in their liabilities rather than aim to maximize yields. Insurers should, for example, match the cash flows of their assets and liabilities to reduce interest rate risks,⁷ for this is the only way for them to invest their assets without threatening the ability to satisfy their liabilities should interest rates vary unexpectedly. Which investment strategy will be optimal depends on the liability structure of the product design. In most cases, the optimal strategy requires risk-free or high-grade corporate bonds for products whose investment risks are borne by the insurers themselves. It can also require investing in riskier asset classes (particularly in the case of certain European products), for which liabilities depend on market returns and policyholders bear some of the investment risk.

Insurers should deviate from this strategy only for specific asset classes in which they can find underpriced investment opportunities over the long term because of distinctive investing skills they possess or market inefficiencies they can exploit. By investing directly in the markets, for example, US insurers can leverage their natural tax advantages to create value that individual shareholders cannot. Take the tax treatment of US corporate and government bonds. Since interest from government bonds isn’t taxed at the state level but interest from corporate bonds is, the market value of corporate bonds reflects a premium of as many as 30 basis points to compensate individual investors for their additional tax burden. Yet in many states, insurers don’t pay tax on interest income—instead they pay a premium tax—so they earn the premium without generating additional tax liability.

You might argue that as investments are moved to lower-risk assets net income will fall, thus reducing earnings per share (EPS). However, the portfolio’s diminished risk level frees up capital to be used by more value-generating activities (such as underwriting insurance risk) and cuts the cost of equity because of the lower market risk exposure. This can in turn raise a company’s P/E ratio and offset the fall in EPS. Moreover, since less capital is necessary to support less risky securities, the reduction in total capital costs can generate an increase in the company’s total value (Exhibit 2).

Give the professionals more expertise

In practice, executing the new investment strategy will require insurers to strengthen the actuarial, risk-management, and portfolio-management expertise of their investment organizations, which have traditionally focused on picking securities. Insurers need a better understanding of the structure and risks of the assets and liabilities they manage in order to choose appropriate GAPM and ALM objectives and processes and to implement hedging strategies for the many risks inherent in their liabilities. For example, the liabilities associated with some product features (such as the variable-annuity guarantees popular in the United States during the late 1990s or the fixed-annuity surrender patterns relative to interest rates) require sophisticated portfolios, since the liabilities don’t respond in a linear way to changing economic factors and can’t be matched perfectly with traditional securities as bonds or equities can.

Some leading North American insurers have responded by adding new teams to support their investment professionals. A few have formed portfolio-management teams and ALM teams within the investment function, as well as derivatives groups to find better hedges for liability risks. Others have started rotating rising stars from the actuarial function into the investment group to ensure the cross-fertilization of knowledge. These players have also built state-of-the-art systems for valuing and tracking assets and liabilities in order to monitor how well they are matched and to test the value of the holdings against fluctuations in the financial markets. In Europe, where a less rigid regulatory framework and the frequent ownership of insurers by banks have led companies to be more sophisticated, Swiss Re, for example, was an early adopter of ALM and other value-based investment practices, well ahead of many primary insurers.

Insurers stand at a critical juncture in shaping their investment strategies after a long-running bull market. They must not only scale back some of the excessive risk that still haunts them but also review their basic investment strategies. In the end, too, they must acknowledge that they can deliver lasting shareholder value only by aligning their product and investment goals more realistically.