In early 2007, the top executives of New York Life, an insurance company, made a game-saving call: with corporate bonds spreads narrowing in a climate of increasing risk, the company boosted the allocation of U.S. Treasuries in its asset portfolio — even though the move would subtract from earnings.
The shift came to be known as the “quality tilt” strategy, referring to a shift toward U.S. Treasury instruments and away from corporate bonds. Five years later, the 12,000-employee mutual continues to lean against the prevailing winds that blow from the rest of financial services, much to its benefit: the company has an “AAA” rating for financial strength, is reasonably well positioned for drastic moves in interest rates, and has managed to maintain a culture that discourages excessive risk-taking.
With decades of experience in insurance, New York Life executive vice president and CFO Michael Sproule is one of the company’s guiding hands in its conservative approach to risk management and its balance sheet. (Like other denizens of the life-insurance industry, probability “tails” and other actuarial concerns are very much at the tip of his tongue.)
For the past two years, Sproule has focused on yet another contrarian initiative: pulling New York Life out of some emerging markets, the same ones other U.S. companies are entering with abandon. Sproule sat down recently with CFO editors to share his thoughts about scenario modeling for tail-risk events, financial industry capital buffers, and what’s wrong with the U.S. economy. An edited version of the discussion follows.
Prior to the financial crisis, New York Life stress-tested the company’s solvency in the event of many extreme scenarios, including the 1918 flu pandemic. Why?
Life insurance is a fat-tailed business in a fat-tailed world. New York Life has been in existence for 167 years and we want to be around for more than that going forward. The reality is that actuaries and risk managers do a lot of very sophisticated modeling to understand the risk profiles of their businesses. The problem is the assumptions that are being fed into these models. Some assumptions, such as those for mortality, are heavily grounded in actual experience. However, other assumptions, such as those for policyholder behavior over a range of interest rates, have no data on which to be based. On top of that, there are all sorts of stress scenarios that you cannot model. So when some companies blow up spectacularly and say, “I was the innocent victim of a one-in-250,000-year event,” I sort of laugh. Because nobody’s models have this kind of precision.
If you want to think about tail-risk scenarios going forward, just think about things that have happened in the last 100 years, which, if they reoccurred, would have a tremendous impact on the life-insurance business. For example, in the last 100 years we had the 1918 flu pandemic. With today’s U.S. population, the same mortality rates from that pandemic would leave 3 million people dead in the United States. That would get everybody’s attention.
In addition, the equity and credit markets would be clobbered. In the last 100 years, we also had the Great Depression, two world wars, the spike in interest rates in 1979 to 1982, and the recent global financial crisis. That is six nonmodelable tail-stress-event scenarios in 100 years. So how uncommon are such tail events in reality? We look to identify the kinds of tail-event scenarios that we think are plausible and stress-test our basic model to make sure we understand, first, the consequences and then, secondly, that we have a plan for how to deal with them.
What other mistakes in risk management, besides having too much faith in models, were made by financial-services companies leading up to the crisis?
There have been far too many risk-management failures. Chief risk officers in so many situations have become chief risk enablers. Long Term Capital Management [the large hedge fund that collapsed in 1998] is my poster child in this area. I like to talk about them because they were the smartest people in any room that you could imagine. They had John Meriwether. They had two Nobel Prize–winning economists. People in the insurance business or the banking business who think they have a smarter group of people managing their risks than LTCM did should think again.
So when it comes to risk taking, I worry about the smartest people in the room because they are capable of justifying incredible amounts of risk taking. Unfortunately, we’ve gotten away from common sense. And keep in mind that LTCM had to be very unlucky not to make a fortune for its principals. We need to figure out how to appropriately compensate people while not creating environments that motivate people to take huge risks with other people’s money. This is often a “heads I win, tails we flip again” construct.
Global financial institutions are arguing against the government imposing stricter capital requirements. But you’re arguing for stiff capital requirements for insurance companies. Why?
If you think about it, if you wanted to destroy the life-insurance business, you might start by getting people who give you real money in exchange for a piece of paper that represents a promise to pay to doubt your ability to make good on your obligations. If that happens, the industry is over. That’s why we’ve been out arguing for stronger, not weaker, capital requirements. And that’s why we hold strong (some might say excessive) capital positions to ensure our ability at New York Life to make good on all of the liabilities that we have created.
We don’t try to game the system. I had the CFO of one of our business units come and see me about six years ago. That CFO had been approached by an investment banker who had a great idea for taking equity investments and restructuring them so that they would basically be a bond with equity-market upside. If you followed your nose you’d say, if I look at this, this is a bond according to the National Association of Insurance Commissioners capital requirements. All that risk-based capital I have to hold because its equity goes away. It’s a structuring device.
So I said to that CFO, “You can go ahead and do that if you want. Pay them their fee. You’ll probably get us some benefit in terms of the cosmetics: we’ll be reporting a higher risk-based capital ratio. But you and I both know that’s not real. So what we’re going to do is allocate internal capital, looking right through this structure, to the reality that it’s equity. So you can go do this if you want, but you’re going to get the internal capital allocated to your unit.” That was the last I ever heard about that idea.
In a move that seems rare in Corporate America of late, you’ve focused on pulling New York Life out of some international markets. Is that because it’s hard to understand the risks in those markets?
No. We just see better opportunities for us to use our capital closer to home. The reason so many life-insurance companies in Europe, the United States, and Japan have been looking for opportunities in emerging markets in Asia is that the growth of the life-insurance market follows the growth of a nation’s gross domestic product. If you look at the underlying fundamentals of the U.S. individual life-insurance market, it’s a marketplace characterized by flat-to-no-growth in life-insurance premiums. Further, fixed expenses are high. That’s not a very good equation unless you have a successful differentiation-based strategy, which we have.
So how do we differentiate ourselves? First, we sell mainly through a proprietary, career-agent distribution channel, while most of our competitors sell primarily through independent agents.
Second, we focus on the middle market. The average-size life-insurance policy that we sell is $300,000. The average-size life-insurance policy that our peers are selling is about $1.2 million. So our peers are focused on more affluent customer segments of the market and are rarely there with us at the point of sale in the middle market. The decision at the dinner table of whether to buy a New York Life insurance policy is, “Do I take the steps to protect my family or do I go and buy the flat-screen TV or take the vacation in Mexico?” So our competition is really not so much our peers.
The markets we want to enter are right here in the United States. So we are really big in the Hispanic market and the Indian market and the Chinese market and the Vietnamese market, and so on. By and large if you’re Hispanic, you’re probably going to tend to buy from an Hispanic agent. Twenty percent of our increase in agents over the last several years has been in the Hispanic marketplace. Our competitors are distributing through independent agents, so they don’t get to control who is selling their products. They really do not have a good way to get to these markets.
Finally, every study says Americans know they are underinsured. The degree of underinsurance actually got worse as interest rates came down. Take somebody who’s making $100,000 a year. They figure they have to leave 10 years’ worth of income to see their spouse and children at least through their children’s college years.
If they think that they are going to earn 8% or 9% on that money, then they don’t have to leave quite as large an estate, because it’s going to grow with interest. But in fact, when you’re down at the interest-rate levels that we’re at now, they actually need to buy a bigger multiple of earnings to create that same income stream for the replacement of those lost wages. That’s what we have going on in the U.S. market today.
What are the challenges in the U.S. economy currently?
At the root of a lot of the global [economic] instability are the major global imbalances. For example, in the U.S. we have the world’s biggest imbalance in terms of the size of the annual current-account and trade deficits we run. In particular, I like to look at personal retail consumption as a percentage of GDP. Currently it is around 72%. If you look at OECD [Organization for Economic Co-operation and Development] numbers, the next-highest country is at 63%. We’re almost 10% of GDP higher on retail consumption than anyone in the Western developed world. For us that’s a trillion and a half dollars a year of excess personal consumption.
Then look at the U.S. savings rate. Our savings rate has declined precipitously since interest rates peaked in 1982. It has bounced off the floor to maybe 4% or 5% right now. And so we have levels of individual savings and consumption that, in my view, are not sustainable. However, we have a government that wants to create an environment to create more consumption, because short term it’s going to be good for the economy. But long term, individuals cannot consume their way out of their problems any more than our government can spend its way out of its problems.