Allison's Ode to the Second Moment


Welcome to the 43rd issue of Allison’s Ode to the Second Moment, a newsletter mostly about pension economics, but more interesting than that sounds.

Pension reporter fetishes

The holidays featured a series of poorly reported stories about pensions. They highlighted many reporters’ inexplicable defined benefit (DB) fetish. As you may know, I am pension-type agnostic. Under certain circumstances, a good DB plan can be better than a good defined contribution (DC) plan. But I’ll take a good, or bad, DC plan over a bad DB plan. If I ever change jobs, I’ll even take a bad DC over a good DB.

But these subtleties are lost in this widely read story in the Washington Post. It takes bending the data to make an argument to another level. Here’s the real deal: DB benefits are expensive because they cover both longevity and market risk. These benefits are expensive to guarantee and there is always an incentive to underfund pensions because the benefits are paid far in the future, when the people who designed these guarantees are gone—just look at public sector pensions.

Many things killed DB plans in the private sector; a main culprit was ERISA, legislation which forced employers to recognize these costs and properly fund DB benefits. Firms weren’t just greedy or screwed by Wall Street. Finally, they had to realize the true cost of their promises.

Another story fetishizes the Dutch plan. I like the Dutch system…I mean who doesn’t? But I am not sure it is fair to compare American DC plans on their own, which are meant to work with Social Security, to Dutch DB plans. In my experience, Dutch CEOs are more paternalistic and willing to take on their employees’ risk.

This story applies DB magical thinking to Social Security by arguing that, if GDP grows at a 2.7% rate, we needn't cut benefits or increase taxes. Sure, people might die early too. But there is only a 3% chance Social Security will work itself out. It is tempting to wait and see if that 3% pans out and keep underfunding, like some pension plans we know. But the problem with the wait-in-see approach is if the other 97% things happen, fixing Social Security will be more expensive than if we fixed it now. It’s like some people don’t understand risk until it is too late.

Turning to better news, there is a nice summary from Jane the Actuary on the state of retirement in America. She says it is not as bad as the hype suggests, but DC pensions (many are poorly executed) raise challenges we must confront. Speak of the devil, here’s an interesting study on why people don’t value annuities.

"Cutting benefits" is the safe-word
I am curious how it will all work out. Many plans are underfunded and cutting benefits is unusual, or even illegal. The state, or corporate, DB benefits individuals actually get aren’t that large. Once people can’t work and are vulnerable, it seems especially cruel to cut their benefits. Older people also have time to assemble and lobby—all of this means pension benefits are hardly ever cut.

Look no further than Greece who, despite some reporting the contrary, barely cut regular pensions (high earners and bonus payments a notable exception) during its debt crisis. The cost of financing pension benefits left little money for programs that help younger Greek citizens. So now young people are moving home to take advantage of their parents’ and grandparents’ higher income. Pensions are a form of national welfare. This is what we in economics call an "inefficient solution."

Things might change. The Economist shows that different forms of pension cuts, mostly those with increasing retirement ages, are more common in rich countries.

Systematic wage risk
I feel like I am being a little harsh this week. The polar vortex makes me grumpy.

Because this story also annoyed me for confusing different types of risk and assuming they require the same remedy. It describes Sweden’s great solution to job loss: retraining and generous unemployment. It works when an individual loses his/her job or even a whole company shuts down. You can pool everyone’s risk through unemployment benefits, but what happens when entire industries disappear and it takes years to find new employment or some people never do? The latter is much more expensive to hedge. I am not sure the Swedish program works as a solution to structural unemployment.

There are things we can do. Chicago Booth collected some good ideas from a wide range of different economists. This is a diverse group and there are some creative solutions on how to revive the middle class.

It’s hard out there for a pension fund manager
A lot of people think a pension fund manager’s job is to score the highest possible return. Some managers behave this way, and this is often how they are judged. Actually, their job is to meet their obligations with as little risk as possible. This is the tension, as more risk increases the odds of paying benefits, but risk also increases the odds of a big short-fall.

Low interest rates make the job harder because managers need to take more risk to make their return targets. The problem is that even stocks are expensive. What to do? Take even bigger risks with alternatives and private markets. There are higher expected returns (allegedly), but also illiquidity and high fees. I am sure that will work out.

Speaking of low rates, my former professor Jean Boivin speculates why interest rates are so low. He cites the usual suspects: low growth and less inflation risk. He also makes an intriguing argument that a riskier environment increased the demand for low-risk assets. He thinks the Asian financial crisis scared investors and governments in Asia and now they’ll pay just about anything for safety. The financial crisis increased risk aversion even more.

Our history is as uncertain as our future
Maybe a few bad decisions in the early 1990s begat the Asian financial crisis, that caused the global saving glut, that set the conditions for the global financial crisis, and that lead to the resurgence of economic populism. We like to view history as a series of causes and effects. We assign probabilities to future events but not the past. Perhaps we should also view history in terms of probabilities and have a deeper appreciation of risk and uncertainty.

Until next time, Pension Geeks!