Insurance regulation, often viewed as a dreary backwater by politicians and economists, is as critical as the banking sector in promoting financial stability, economic growth, and consumer protection. Life insurance and pension plans are almost as ubiquitous as mortgages. Their providers hold $50 trillion worth of assets worldwide.¹ In the last chapter, I argued that financial stability is best achieved through a transfer of risk based on the different risk capacities between short-term funded institutions like banks and long-term funded institutions such as life insurers and pensioners. Banking and insurance stability are simply different sides of the same coin. To view them as two separate endeavors is a grave mistake. The regulation of both banking and insurance must be integrated from a systemic risk perspective.

For the moment, we will leave the institutional aspects of integrating banking and insurance regulation and return to them in Chapter 13. This chapter takes a deeper look at the alternative regulatory framework proposed in Chapter 5 and peers at it from the perspective of life insurers and pension funds rather than just banks. If I expressed frustration in the previous chapter that the banking regulation train had left the station in the wrong direction, I would like to express urgency here, because at the time of writing the regulation train has only just pulled in to the insurance platform. There is still a chance for regulators to fix what has been proposed before it is too late. Consequently, we will also take the opportunity to see how our alternative framework ­differs from the current framework of regulation of insurance firms.

Solvency II² is an EU directive designed to harmonize the regulation of insurers and pension funds. However, it has significant extraterritorial reach beyond the EU for reasons peculiar to it and to insurance regulation worldwide. Seeking greater consumer protection and an improved financial system, the European Parliament approved the Solvency II Directive in March 2014. It is slated to come into effect on January 1, 2016.³ In its current form, the asset allocation that Solvency II imposes on firms will be a disaster. It will be a catastrophe for consumers, financial stability, and economic growth. An alternative asset allocation, attuned to the risk that the assets of an insurer fall short of its obligations as they come due, would correct this problem. Instead we have been presented with one that is overly sensitive to the current volatility of asset prices.

On paper, Solvency II is not an international standard of insurance regulation, like Basel II is for banking supervision. However, Solvency II will easily assume as important an international role as Basel II. Its significant extraterritorial reach emanates from three separate directions. Since Solvency II is designed to cover an insurance company operating across different EU countries, the focus of its supervision is set “upward” to the parent company and not downward to the level of the local entity or product. Consequently, the US subsidiary of insurance companies headquartered within the EU, such as Allianz, Aviva, ING, and Mapfre, fall within the scope of Solvency II compliance.

The EU subsidiary of a non-EU insurance group like MetLife, Canada Life, Travellers, or Tokio Marine will also have to comply with Solvency II as stand-alone entities.⁴ In a partial attempt to address the reporting, accounting, capital, and other issues arising from this extraterritoriality, intense international efforts are underway. The EU Commission and the US National Association of Insurance Commissioners (NAIC), have been meeting to ensure consistency in the regulation of EU and non-EU group insurers. US insurers are particularly angry about Solvency II’s reach and the prospect of an additional, and in their eyes unnecessary, layer of regulation. Although the US insurance market is bigger than the EU’s because US insurers are regulated at the state rather than the federal level, the EU is the larger insurance jurisdiction. Consequently, obtaining the status of regulatory equivalence in order for there to be only one set of regulations, often requires US States and the rest of the world to shoehorn their domestic regulation into the requirements of Solvency II.

Separately, in September 2008, prompted by the collapse of US-headquartered insurer AIG, the Financial Stability Board (FSB) began efforts to create an international standard for the regulation of what it considered to be systemically important insurance companies.⁵ This list included a number of non-European groups like AIG, MetLife, and Prudential Financial in the United States and Ping An Insurance in China.⁶ However, given the EU countries influence on the FSB, the long reach of Solvency II, and the lack of fundamental thinking about the nature of insurance companies’ risks, a version of Solvency II has been established as the standard to which the FSB holds these firms.

Regulation of banks and insurers has long pivoted around the idea of moving firms toward best practice.⁷ Basel I and II, designed for only the internationally systemically important banks, became the standard for all banks. Deviations were considered suspect. Pushed along by extraterritoriality, the pressure for equivalence, and the FSB’s initiative, Solvency II, with its approach of “market consistent” valuations and risk assessments when calculating capital requirements, has become the “best practice” standard in insurance and pension fund regulation globally. It is the embodiment of the current approach to insurance regulation and the current path regulators are on, the world over.

What’s Wrong with the Current Approach to Insurance Regulation?

Solvency II was developed at the turn of the century and modeled on the Basel II Accord, which regulates the supervision of international banks. Both documents reflect deep faith in the “marketization of finance”. As discussed earlier, this is the idea that markets are better than financial firms and individuals at pricing and managing financial risks. This belief in markets, as discussed in Chapters 4 and 5, was not limited to regulators but was part of the larger zeitgeist of the 1990s and a common conviction in the windward side of financial booms. Underpinning the marketization of finance were the related beliefs that all assets have one price—the price that can be obtained were the asset to be sold in the marketplace—and that the riskiness of an asset is the short-term volatility of this price. Remember that the capital-adequacy regimes developed under both Basel II and Solvency II marched in union with the shift toward mark-to-market accounting and risk management systems based on the volatility of current prices.⁸ These ideas followed the capital asset-pricing model popular with mutual funds investing in continuously trading markets.

The majority of assets in the world, such as property, infrastructure, and human capital, are not continuously traded. Yet many regulators saw their task as shining a bright light on the dark world of insurers and pension funds buying unlisted and untraded investments. They sought to create a brave new world where life insurers would hold assets that everyone could price and assess the risk of in the same way. Prior to the GFC, there were many perceived benefits to such an approach, including creating fraud-busting transparency, leveling playing fields, and the imposition of market discipline. Some of the frauds I discussed in earlier chapters brought these issues to greater prominence in the past than they may be today. Investment banks also fancied this approach as it opened up an entirely new universe of financial innovation where risks and balance sheets could be sliced and traded.⁹

Solvency II’s solvency capital requirement is made up of a series of capital requirements for the risk of different activities, including, insurance, counterparty and investment risk. In the last quantitative impact assessment reported by the European Insurance and Occupational Pension Authority (EIOPA), capital for the market risk of investments constituted the largest component of life insurers’ capital requirements under the new Solvency II proposals. Solvency II requires firms to set aside an amount of capital for market risk that would offset a fall in asset values over one year of a size so large that it would only occur once every two hundred years.

With equities quoted on an EEA¹⁰ or OECD exchange, the standard formula of Solvency II requires capital provision for a 39 percent fall in prices. For other equities like emerging-market equities or developed-country private equity, it requires provision for a 49 percent¹¹ fall. There are adjustments to take account of the financial cycle, the tax effects of insurance loss, and the risk-absorptive aspects of some technical-insurance provisions. However, according to quantitative-impact assessments, insurers would still need to put up capital in the order of 23 percent–28 percent of the value of equity holdings. By comparison they are only required to put up 3.0 percent of the value of a 10-year A-rated corporate bond or 2.1% of the value of an EU government debt instrument with an A rating. Preparation for Solvency II and previous regulatory preferences have already pushed insurers’ holdings of equity and property down to half of what they were in the 1990s. However, despite being just 12 percent of all of their holdings today,¹² these holdings will still account for 37 percent of life insurers’ capital-adequacy requirements under Solvency II.¹³

Parallel with Basel II, large insurance companies may adopt regulatory-approved internal risk models beyond these standardized specifications. But these models are not a license to deviate from the central treatment of risk as a once-in-a-two-hundred-year annual fall in asset prices. It is merely a license to extend this approach to asset classes not yet considered by the regulators in the standard formula. Following a series of quantitative-impact assessments and simulations, it is widely accepted that, as a result of the disproportionate impact on their after-capital-charge returns, Solvency II will lead to a switch out of public and private equity, infrastructure bonds, property, and low-rated corporate bonds.¹⁴ Regulators do this in the name of protecting consumers from insolvent insurers. Yet it is actually not in the interests of consumers of long-term insurance or pension products and has other, wider, adverse consequences for financial stability and economic growth. It is hardly in the insurer’s interest to annoy their regulator, or even to worry about systemic risk. However some insurance CEOs have seen fit to draw attention to the likely negative impact on infrastructure financing in both private and public remarks.¹⁵

The architects of Solvency II describe their framework as being risk sensitive and adhering to market-consistent valuations. These terms sound so terribly sophisticated and self-evidently sensible that the underlying thinking behind Solvency II is afforded scant scrutiny. After all none would instead insist on being risk insensitive or market inconsistent. Yet this framework rests on a fundamentally flawed view of the investment risk of a life insurer or pension fund. The riskiness of the assets of a life insurer or pension fund with liabilities that will not materialize before ten, even twenty years, is improperly measured by a once-in-a-two-hundred-year drop in prices on a one-year basis. The capital asset-pricing model fails to take into account that institutions with different liabilities have different capacities for absorbing different risks.

Of course not all insurers have long-term liabilities. Casualty insurers, like those who write motor or health insurance policies, have potentially short-term liabilities. However, in Europe and elsewhere, 80 percent of the assets held by insurers are held by life insurers.¹⁶ The regulatory-induced reduction in the holdings of long-term investments by long-term savings institutions will lead to a fall in returns to and a rise in risks taken by consumers. It will reduce the risk-absorptive capacity of the financial system, making it less resilient. Anticipation of Solvency II and other regulatory and accounting pressures has already reduced their equities holdings but today insurers and pension funds still hold approximately 15–20 percent of equities in developed markets.¹⁷ The arrival of Solvency II will accelerate this trend of equity disposals and increase the cost of long-term investment by companies, in turn leading to a stunting of economic growth. To understand fully why it is not in the interests of consumers, the financial system, and the economy for life insurers and pension funds to eschew long-term assets like equities, we must do a quick rehersal of the fundamental principles of risk capacity introduced in Chapter 5. We will then apply these principles to the assets of a life insurer or pension fund with 20-year liabilities.

The Fundamental Principles of Investment Risk

The risk free rate of return is that available from taking no risk—such as cash in a bank with a US government deposit guarantee. An investment return over and above the risk-free rate of return can only be obtained by taking an investment risk. Different potential returns are earned by taking different risks. The main investment risks are credit, liquidity, and market risk. For example, if an investor in a US corporate bond made a 7 percent return over a year, this return is composed of:

1. The risk-free rate (the return available on cash); plus
2. The return from taking a credit risk (the risk that the issuer of the bond will go bust over the year); plus
3. A liquidity risk (the risk that the holder of the bond will have to accept a lower price if he has to sell it quickly and cannot wait to find a more interested buyer); plus
4. A return for taking a market risk (the risk that the price of the bond falls because the market value of the cash flow declines, perhaps as a result of rising US interest rates).

Investment risks are categorized into credit, liquidity and market risks because they are fundamentally different rather than as a matter of convenience. The test of this difference is that if an investor did not want to take one of these risks, each of these three risks that she did not want to take would have to be hedged¹⁸ differently.

How to Hedge Different Investment Risks

Hedging credit risks is done through diversification. For example, to spread the credit risk of holding a General Motors bond, it can be held within a portfolio of bonds issued by its main competitors, like Chrysler and Toyota, who might sell more cars if GM were to get into trouble. It could also be held with bonds issued by companies whose success may come at the expense of all traditional car companies—like Telsa, an electric-car maker; BP, an oil company; Bombardier, a manufacturer of high-speed trains; or Microsoft, the owner of Skype, favored by telecommuters.

Liquidity risks are best hedged by having time to sell. An illiquid asset is one whose price falls below what could otherwise be obtained if there was more time to find a suitable buyer. My glass-and-steel house in a row of Victorian terraces in London will only appeal to a narrow band of buyers. If I was forced to sell tomorrow, it would fetch a far-lower price than if I could wait for that person with similar taste to find it. One way to create more time to hedge this liquidity risk is to finance the asset with a long-term mortgage. In general, the more time you have to wait for a suitable buyer, the less liquidity risk you carry.

Market risks are hedged through a combination of diversification across uncorrelated market risks and with time to allow for panic, uncertainty, or stretched valuations to unwind.

What is clear is that liquidity risks and credit risks are hedged in distinctly different ways. Time hedges liquidity risk, but more time in which a shock can arrive and a company can go bust increases credit risk. There is little uncertainty and credit risk in holding a General Motors bond for one day but much more if it is not sold for 20 years. Diversification across similarly sized but differently correlated credit risks reduces aggregate credit risk. However, diversifying across equally illiquid assets does not reduce liquidity risks. Credit and liquidity risks are different and cannot be meaningfully added up or sliced and mixed together despite the best efforts of former Russian nuclear physicicsts working at US investment banks at the turn of the century or anybody else.

Different Capacities to Hedge Different Risks

Financial firms or individuals have a natural capacity to hedge one or more types of risk. A bank with thousands of borrowers in different economic sectors has an innate capacity to diversify its credit risk. It can get paid for taking credit risks and self-insure against these risks through diversification. That is the real business model of banks. However, a bank with loans funded by cash deposits or money market funds that can be withdrawn overnight, has a limited natural capacity to hedge liquidity and market risks. Ideally such a bank should earn its credit-risk premium by lending to a diversified group of borrowers and it should charge a goodly sum that covers the cost of transferring its portfolio of market and liquidity risks to someone else who can hedge these risks more cheaply. A contributory factor to the GFC was that risk transfers went in the opposite direction to this ideal. Regulatory capital charges on credit risks but not liquidity risks incentivized banks to sell credit risks and buy liquidity risks. For example, where banks sold credit risks through a special purpose vehicle, they often sweetened the deal with a liquidity backstop—an agreement to buy back the assets.

When Lehman Brothers was forced into bankruptcy on September 15, 2008, creditors of the UK arm contemplated losses in the order of $200 billon as the price of Lehman’s illiquid credit assets plunged below its liabilities.¹⁹ However (and not without irony), the longer the administrators took to unpick the complexity of the banks’ assets and liabilities, the further the prices of those assets recovered. Seven years later, the administrators of the UK operation were able to announce that they had recouped all of the cash owed to secured and unsecured creditors. Despite their own enormous bill, a modest surplus was left.²⁰ Illiquidity played a major role in the last banking crisis just as it had in many before.

A life insurer or young pension fund with a likely concentration of payouts in 20 years time has a natural capacity to take liquidity and market risks and earn the associated risk premia. It does not have a natural capacity to hold credit risk.²¹ Imagine two different portfolios of assets with the same expected return over the next 12 months. The first is a package of poor, but highly diversified and liquid credit risks, listed and traded frequently on an exchange. The second is a portfolio of government-guaranteed loans that cannot be sold on without a lengthy permission process. This second instrument has low credit but high liquidity risks. Imagine, too, that the investment strategy of the life insurer is to lock away the two portfolios in a safe and only open it when a life-insurance payout in made in 20 years. While the average return maybe the same, the distribution of outcomes of these two portfolios change with time. There is a higher likelihood that the first portfolio with poor credits underperforms the second portfolio with poor liquidity because over a significant time period one of the credits in the portfolio of poor credit risks will have gone bust. The longer the period, the greater the probability of this outcome.

The Challenge for Long-Term Investors Is Short-Fall Risk, Not Short-Term Volatility

The risk that matters to the life insurer or pension fund is the risk of a shortfall in the return of the asset when they need it. If they have a liability in 20 years, taking a liquidity risk during the first 10 years does not engender a shortfall, but taking credit risks during that time does. Shortfall risk is different and not well reflected, if at all, in the daily, monthly, or even annual volatility of the price of the asset. The capital-asset pricing model is not designed for someone facing shortfall risk in 20 years but rather for the investor who may have to liquidate her assets in the short term. Assets with low annual volatility, but where the risk of a loss rises over time, or cannot be reduced with time, may pose greater shortfall risk for a long-term investor than assets that exhibit high annual volatility but with risks that fall over time. The same amount of cash invested in a diversified portfolio of liquid credit risks may have a lower annual volatility but a higher risk of failing to achieve the investment objective after 20 years than a diversified portfolio of illiquid private-equity assets. From a life insurer or pension fund’s perspective, Solvency II gets the notion of risk completely muddled.

The best strategy for investors is to first hold the risks that they have a natural ability to hedge and sell the ones they cannot hedge to those who can. The risk inherent in a particular asset is not the same for everyone at all times. Risk is contingent on who owns it and for what purpose. To satisfy a hunger for investment returns, begin by eating the free lunch on offer and understand that all subsequent lunches come with a bill. This may seem obvious, but Solvency II actually generates the opposite behavior.

The S&P 500, Life Insurers, Pension Funds, and Solvency II

The investment principles discussed can be tested by examining the risks faced by a US subsidiary of a European life insurance firm when holding the components of the S&P 500 Index of large, liquid US stocks. Since 1928, the S&P 500 Index has boasted an average return significantly above the risk-free rate of US government T-bills or the riskier 10-year Treasury bonds.²² The credit and liquidity risks of this index²³ are small so this extra return is likely compensation for market risk. Recall that market risk is hedged by both time and diversification. The longer the time period over which the S&P 500 stock portfolio is held and the more the market risk is spread, the greater the probability that returns at the end of the period will be positive and the shortfall risk reduced. Let us examine how this has played out in the past.

Over the last 22 years, the average daily return of the S&P 500 has been 0.03 percent.²⁴ While this is more than double the average daily return of three-month US T-bills, this figure is not properly representative of the distribution of daily returns. Forty-eight percent of daily returns were negative. Six percent of daily returns were more than two standard deviations south of zero²⁵—reflecting a “fat-tailed” distribution where the likelihood of extreme outcomes is greater than a normal distribution.²⁶ A money market fund or casualty insurer that may have to liquidate assets on short notice has no time to spread this size of market risk. The additional return for that investor from owning the S&P versus T-bills is not a free lunch. It comes with a significant increase in shortfall risk.

Over the past 86 years, the average annual return of the S&P 500 has been 12 percent which is no less than the daily return annualised.²⁷ However, the risk or distribution of these annual returns is meaningfully different than for daily returns. Only 29 percent of annual returns were negative versus 48 percent of daily returns. A mere 3 percent of annual returns were more than two standard deviations below zero versus 6 percent of daily returns. The distribution of returns is less fat tailed and significantly shifted into positive territory.

Since 1928, the average cumulative return over discrete 10 year periods has been 196 percent. Each of the 10-year periods except for 1928–1938 was positive. Although this single losing decade included the 1929 Great Crash, the Great Depression, and the 1937 stock market collapse, the loss over ten years was only 6 percent.²⁸ Similarly, only 7 percent of the 76 overlapping 10 year periods since 1928 have been negative. This confirms the earlier proposition that liquidity and market risks fall with time. The shortfall risk for an investor with long-term liabilities is lowered by holding assets with low credit risks but high market and liquidity risks.

The picture is stronger still if we look at 20 year time spans—though there are only a few of these discrete periods. The average return is 583 percent, or 370 percent if you remove 1938–1958 on the grounds that the rebound from the Great Depression to postwar euphoria is unlikely to be repeated. None of the four discrete 20 year spans or even the 66 overlapping 20 year periods have been negative.²⁹ There have not been any negative returns either from investing in 10-year Treasury bonds over the same 20 year spans. However, the average 20-year return for 10-year US government bonds has been much lower than for equities at 165 percent (or 120 percent if you remove the best decade, 1988–2008, on the grounds that the great disinflation is unlikely to be repeated from here).

When considering the shortfall risk of assets backing long-term liabilities such as life insurance and pensions, we should consider the impact of inflation on long-term returns. Over 20 years even modest rates of inflation can seriously erode the purchasing power of a pension. Consumers are more concerned with the real value³⁰ of a future pension or life-insurance payout than they are with its nominal value. If we were to consider real returns over 10 or 20 year holding periods, equities might be safer and more rewarding than government securities. Since 1928, the average real 10-year cumulative return on equities has been 113 percent with only one decade from 1968 to 1978 of negative real returns of –17 percent. Over the same span, the average real 10-year cumulative return on US government bonds has been 29 percent—just one quarter of the real equity return —with two decades of negative performance when they were –13.7 percent (1938–1948) and –3 percent (1968–1978).³¹ There have only been nine discrete 10 year terms since 1928, but, if we consider overlapping 10 year windows, equities have had a positive real return 89 percent of the time and US government bonds only 66 percent of the time.

Now recall that under Solvency II, life insurers are required to hold capital of up to 24 to 28 percent of the value of an equity portfolio and little or no capital for holding a government-bond portfolio that prior to the European credit crisis might have contained Greek and Cypriot bonds. Unlike a money market fund, a life insurance or young pension fund (with a high probable concentration of payouts in twenty years’ time and a low probability of payouts in the short term) does not need to liquidate the vast majority of its portfolio in the short term. It can earn the substantial market-risk and liquidity-risk premia available in moving from bonds to equities without substantially increasing shortfall risk. If we consider inflation, there may well be an increase in real returns with no increase in short-fall risk at all from the switch.

Consumer Protection

When presented with the theory of investment risk as argued, some regulators have pleaded that it is better to be safe from a practical perspective than waylaid by fanciful theories. John Maynard Keynes once said that practical men who believe themselves exempt from any intellectual influence are usually slaves of some defunct economist. In this case, they are the slaves of a defunct financial theory that equates all financial risk to annual value-at-risk estimates—an idea that proved wholly inadequate during the last crisis.³² Even worse, here playing it safe means shifting risk to those least able to bear it—the consumers of insurance and pensions.

The wider consequence of forcing life insurers and pension funds to hold liquid assets when they do not need liquidity is that the same amount of life insurance or pension is then more expensive for consumers to buy. Insurance is a luxury rather than a necessary good. Its consumption rises with income. If we add that most ordinary consumers are liquidity-constrained, should the same amount of future life insurance or pension costs more today, the average consumer will buy less rather than sacrifice the grocery bill. They are likely to be less insured. Solvency II is shifting risk from those most able to manage and bear it—institutions with full-time professionals that pool and spread risks—to those least able to manage and bear it. Although this book is focused on systemic risks, there are deep connections between systemic risks and consumer protection and this is but one example of where getting systemic risks wrong, often in the name of consumer protection, can harm consumers indirectly.

Regulators are not only pushing life insurers into an asset class with generally lower returns they are doing so at a time when the outlook for long-term bond returns is particularly skewed to the downside. Near-zero interest rates, modest economic growth, postcrash risk aversion, geopolitical uncertainty, and a massive central bank bond-buying program have all pinned bond prices to the ceiling. A whole constellation of stars will have to be correctly aligned if prices were not to fall sharply. Basel II capital-adequacy rules helped push banks into excessive mortgage risk. Today, Solvency II is leading consumers of insurance companies like lambs to a slaughter.

Systemic Risks

It is right from an investment perspective for life insurers or young pension funds to hold good-quality credits with high liquidity and market risks–like diversified portfolios of government-backed infrastructure bonds, asset-backed securities or public and private equity. It is also correct from the perspective of the insurance buyer who gets more coverage for a given premium. And it is proper too from a systemic-risk perspective. If these market and illiquid risks were held by short-term investors, like money market funds, hedge funds, or banks, there would be phases of steep, self-feeding declines in asset prices as all investors try to offload the same assets simultaneously whenever liquidity or market conditions went south. Doing so pushes up short-term volatility and reported risk, promoting more sales and price declines. Putting long-term risks in fast hands will make the financial system more fragile.

The life insurer or pension fund unpeturbed by short-term volatility would be the ideal counterparty for a bank wanting to sell a package of market and liquidity risks where the bank retains the credit risks. There are clear individual and systemic benefits of this type of risk transfer. Too often in the past, the precise opposite risk transfer took place because transfers were not driven by where the greatest risk capacity was to be found but where the lowest regulatory-capital charges were. Life insurers were seduced by the low capital charges incurred when buying packages of securitized credit risks with high credit ratings. AIG, the insurer, did so in bankruptcy-inducing proportions, but other insurers were also guilty of similar conduct if on a more modest scale.³³

Arguably, there is no reasonable amount of capital that could bring safety to a financial system if risks are held in all the wrong places. We are doomed to failure if all the liquidity risks are held by those with no liquidity and all the credit risks are held by those who cannot diversify them. From a regulatory perspective, risk capacity is a more critical concept than the “market consistent valuations” and “risk sensitivity.” It is a grave disappointment that it remains largely neglected by regulators.

Economic Growth

Critical to any economic growth is investment. We need capital expenditure on laying fiber-optic cables; launching communication satellites; and building airport, train, and road networks. All else being equal, the lower the cost of capital, the higher the level of investment. Investment assets tend to have long development times as well as being large and not easily divisible. Half a train station is not as easy to sell as a completed one. These projects carry huge liquidity and market risks. The funding costs would be lower if they were financed by those with a natural capacity for taking liquidity and market risks such as the life insurers and pension funds. If these natural buyers of long-term assets are denied the ability to do so, then the cost of capital will be higher than it would otherwise be with the resulting effect of lowering investment and economic growth.

The degree to which investment would be lower than it could otherwise be depends on the amount of long-term savings versus short-term savings and the degree to which short-term savers can substitute for long-term savers in holding long-term assets. In Europe, life insurers and pension funds own approximately €10 trillion of assets, or more than half of EU GDP or half of all institutionally owned assets. If Solvency II did not get in the way, a life insurer with 20-year liabilities would always be able to outbid a hedge fund that offers near-immediate liquidity to its investors because the insurer does not need to find a costly hedge for the liquidity risks. These are not easily substitutable investors. Consequently, investment will suffer at a time Europe can least afford for it to do so.

What Is to Be Done?

The fundamental principles of risk capacity previously discussed point to a different but straightforward approach to the solvency-capital requirement. It is an approach that reflects the risk that the investment will fall short of what is required. The shortfall risk of an asset is not independent of the liability it is up against but rather is intrinsically linked to it. If the liability is an expected payout of €100,000 in 10 years and the asset is a monthly investment in a basket of public equities, by using past data on the 10-year returns of the asset, we can estimate the likelihood that the investment will fail to achieve the payout. Capital requirements would be sized to offset this risk.

This would imply very different capital requirements from those of Solvency II. For instance, consider a firm with long-term liabilities. The capital to be set against a diversified portfolio of blue-chip equities might be less than for a portfolio of long-term, liquid bonds that carry the same expected annual return. This need not be done one liability at a time. It can be done with buckets of similar liabilities and buckets of assets matched against them. Where the shortfall risk is greater than, say, 0.5 percent, the insurer would be required to reduce it by raising the level of the premium, changing the asset class, or raising the level of capital.

Conclusion

To consider the risk of a life-insurance and pension fund as being well measured using an estimated once-in-two-hundred-year decline in annual asset prices is fundamentally flawed. The return and price of assets already reflects a measure of the asset risk to the average holder. Consequently, the starting point for the regulator ought to be whether the risks to life insurers or pension funds from holding an asset are lower or higher than already reflected in the market price. In both this chapter and the last we have shown that this relates to what the asset is being used for and by whom. Market-consistent valuations sound sensible but are not a desirable objective across individuals or firms with different liabilities. Private-equity funds are a risky investment for a casualty insurer to hold but safer for a life insurer with liabilities beyond the redemption period of the fund. To the life insurer, it matters not what the price of assets and the risk of holding those assets is tomorrow or at year end. Their main concern is the risk that their investment return falls short of what is required when it is required: shortfall risk.

The consequence of properly matching risk taking with risk capacity is that insurers, consumers, the wider financial system, and the economy will be able to move to a superior risk-return point. If Solvency II were to appropriately set solvency-capital requirements for investments around the shortfall risk of holding certain assets against certain liabilities, consumers would have cheaper but equally adequate insurance. The financial system would be safer as risks flowed to where they were best spread or diversified, and the resulting lower cost of capital would boost investment and economic growth. Nothing stops us traveling in this direction but ourselves. Regrettably, in its current form, the proposed capital regime in Solvency II will take us the opposite way.

___________________

¹Bank of England, “Working Party on Pro-Cyclicality” (London: Bank of England: June 2014).

²Solvency II Directive, European Commission 138 (2009). See also Commission Delegated Regulation 2015/35.

³Solvency II was initially due to be implemented on January 1st, 2013, but concerns over its impact caused that date to be pushed back. Even post-January 1, 2016 its implementation will only be phased in.

⁴There were early concerns that the whole group of a non-EU insurance company with an EU subsidiary would have to comply but that is not the case. US and other non-EU groups complain that having the EU subsidiary comply with Solvency II as a stand-alone entity, with its own capital-adequacy requirement, reduces the scope for its customers to benefit from efficiencies in capital management of the larger group.

⁵Whether insurance companies are systemic in the same way as banks is an interesting question but space constraints preclude this from being properly addressed here.

⁶The FSB has declared the following insurance companies to be globally and systemically important: Allianz SE American International Group Inc., Assicurazioni Generali SpA, Aviva PLC, Axa SA, MetLife Inc., Ping An Insurance (Group) Company of China Ltd., Prudential Financial Inc., and Prudential PLC.

⁷See Charles Goodhart, The Basel Committee on Banking Supervision: A History of the Early Years, 1974–1997 (Cambridge, UK: Cambridge University Press, 2011).

⁸Value-at-Risk (VaR) and Daily-Earnings-At-Risk (DEAR) are examples of these risk management approaches.

⁹It was not simply good luck for investment banks that regulators adopted this approach. They had a strong hand in making the case. See Avinash Persaud, “Banks Put Themselves at Risk in Basle,” Financial Times, October 16, 2003.

¹⁰The European Economic Area (EEA) is made up of the EU member states plus Norway, Liechtenstein, Iceland and, pending ratification, Croatia.

¹¹See EIOPA’s website for further details: www.eiopa.europa.eu

¹²Ibid.

¹³Ibid.

¹⁴Even before the most recent quantitative-impact assessment in 2014, the Economist Intelligence Unit Report for BlackRock conducted a survey looking at 223 insurers with European operations, finding that 97 percent of insurers agree that the equity-risk premium would have to rise to justify them investing in equities given the new capital charges. Ninety one percent agree with the idea that share prices will be lower as a result of Solvency II, and 91 percent agree that corporations will respond by switching from equity to debt issuance. For a simulation exercise on the impact of Solvency II on insurers’ investments, see Andre Thibeault and Mathias Wambeke, Regulatory Impact on Banks’ and Insurers’ Investments (Ghent, Belgium: Vlerick Centre for Financial Services, September 2014).

¹⁵On August 14, 2013, the UK’s Independent newspaper, quoted Mr. Tidjane Thiam, then CEO of Prudential Insurance of the UK, as saying that “the proposed Solvency II regime could prevent insurers from investing in infrastructure and property . . . costing the UK jobs and growth.”

¹⁶EIOPA

¹⁷See Bank of England, June 2014.

¹⁸Hedging of risks means neutralizing or offsetting them so that they no longer have impact. Individuals often hedge risks, like the risk of a motor accident, by buying insurance against that risk.

¹⁹See William R. Cline and Joseph E. Gagnon, Lehman Died, Bagehot Lives: Why Did the Fed and Treasury Let a Major Wall Street Bank Fail? (Policy Brief 13-21, Peterson Institute for International Economics, September 2013).

²⁰According to Tony Lomas, the Pricewaterhouse Cooper partner leading the administration, as reported in Lehman’s UK Unit Administrators Foresee £5bn Surplus, Financial Times, March 5, 2014.

²¹This is one reason why it is inappropriate for these companies to buy bail-in bonds issued by banks; see Avinash Persaud, Bail-In Securities Are Fools’ Gold (Washington, DC: Peterson Institute for International Economics, November 2014).

²²Although the S&P 500 is estimated back to 1928, the original index started form in 1928 as the S&P 90 until 1957 when it became the S&P 500.

²³This is through the futures market. Futures in the S&P 500 (as opposed to funds that hold the component shares) are one of the most liquid financial instruments.

²⁴Data was provided by CLSA Ltd.

²⁵Standard deviation is a measure of the distribution of outcomes. In a normal distribution—one that looks like a bell jar—68.3 percent of outcomes are within one standard deviation from the mean, 95.5 percent of outcomes are within two standard deviations from it, and 99.7 percent of outcomes are within three.

²⁶Ibid.

²⁷See data of the Federal Reserve database in St. Louis (FRED), http://research.stlouisfed.org/fred2/

²⁸Returns are calculated with re-invested dividends. The peak-to-trough decline during this ten year period was greater than zero but this highlights my point that more time reduces market and liquidity risk.

²⁹Even the harrowing period in the stock market from 1929 to 1948 is positive once dividends are reinvested. Today, in the US, where dividends are becoming a rarity, the importance of dividends on equity returns in the past is often forgotten.

³⁰The real return of an investment, mean returns after inflation has been subtracted.

³¹Author’s calculations derived from the FRED data.

³²See Avinash Persaud, “Market Liquidity and Risk Management” in Liquidity Black Holes: Understanding, Managing and Quantifying Financial Liquidity Risk (London: Risk Books, 2002).

³³We all tend to fight the last war and avoiding a repeat of AIG has been a strong motivation for insurance regulators in recent years. But, arguably, AIG was a special case of an institution acting like a bank while in an insurer’s clothes and could be dealt with more specifically.