Optimal portfolios for retirement

You are building up assets for retirement. What is the best portfolio to do this? This question has been studied before, but has neglected flexible labor supply. Indeed, if for some reason your retirement did not work out as expected, you should still have the option to work (again/more). Gomes, Kotlikoff and Viceira takes this idea to a calibrated life cycle model and come to the following conclusions.

First, no surprise, young people should start accumulating for retirement right away. But the accumulation of assets stops before retirement. Why? As wages decline typically after age 55, it is better to increase leisure and start drawing on assets.

Second, young people should invest fully in stocks and take advantage of the higher return in the long run. In the early thirties, they can start investing in bond, at an increasing share all the way to retirement. But the stock share always stays above 45%.

Third, once retired, draw first on the bonds, so that the stock share increases. Why would one want to increase the risk of the portfolio as one ages? It is not that risky, as there is fixed income from social security.

And what did flexible labor supply add to this? One needs to accumulate less assets for retirement, the self-insurance aspect being lessened. Also, it is possible to have a more risky portfolio and thus reap the benefits from the long-term equity premium over bonds. Finally, the modeling indicates that one should retire gradually. The only sudden shift in behavior is due to social security payments kicking in at 65.

Why three medals at the Olympics?

It is interesting to follow Olympics in various countries. While everywhere there is focus on domestic athletes, big countries, say the US, only consider gold medals to be worth mentioning, while in small ones, say Denmark, even Olympic diplomas (ranks 4-8) make major news. This brings up the question: why reward more than the winner?

Pavlo Blavatskyy demonstrates that giving a single prize elicits a lot of effort from a few athletes. Giving several prizes reduces the effort of those few, but this may increase the overall effort as more athletes participate.

Given, as I suggested above, only gold medals are worthwhile in big countries, I wonder whether we can see more effort from the top athletes from big countries, while we see more effort from small country citizens among diploma-worthy athletes. An empirical test of this would be difficult, though. Indeed, a prediction of this theory would to find proportionally more athletes from small countries in ranks 2-8. But larger country athletes could simply be missing from because a fellow citizen won.

Who will win the Olympics?

Now that the Olympic Games are about to start and the patriots start counting medals, let us ask what determines the success of a country at these events? Obviously, size matters, in particular for team events (and this is why I frustratingly root for the little countries). Bernard and Busse show that economic development also matters. This is not surprising, as a country need to reach some development level to devote resources to sustain a competitive sports program.

Lui and Suen confirm this. Kuper and Sterken as well and argue that the home field advantage has become less important over time. Glen Roberts argues that a cold climate helps as well. Moosa and Smith add health expenditures. Stefan Szymanski claims that medals counts are very predictable and that any discrepancy should be attributable to cheating, an argument that Jacob and Levitt have used successfully in other research.

Of course, Olympic success can be a function of how much sports and in particular competitive sports enjoy public support. Rathke and Woitek make exactly this argument using stochastic frontier analysis which measures how far a country is from its potential. Tcha and Pershin find that, much like in international economics, high-income countries specialize less in particular sports.

Gambling for savings

We all agree that playing the lottery is a very poor way to save for retirement. One can rationalize it under some circumstances, but it is generally frowned upon. It turns out that 40 percent of UK citizens play a lottery to save for retirement. And this is encouraged by the government.

They invest in Premium Bonds, a financial vehicle known around the world as lottery-linked savings (LLS). LLS programs are especially popular among low-income households and, in the case of developing economies, with people outside the banking system. How do they work? Bonds are issued that yield a rather normal return, but periodically a lottery is drawn among all bonds yielding some additional return, sometimes substantial. So it is like buying a bond and a lottery with the same ticket.

Do people who buy LLS bonds buy them for the savings or for the gambling? Peter Tufano answers this question by using time series of Premium Bond sales in the UK, comparing them to various betting schemes, other ways of saving, economic activity and marginal tax rates (PB winnings are not taxable). The outcome: the size of the largest prize matters, even though it represents only 2% of the expected return. PB sales respond positively to overall gambling activity, but also to relative yields with other financial vehicles. But the terms of the PBs do not affect their redemption, indicating that they are mostly a regular savings vehicle.

Are LLS bonds a good way to introduce low-income households to savings? The British experience seems to indicate so. But they are prohibited in many countries, including the Unites States, as they violate lottery laws.

The careers of the educational elite

The life cycle, the educational outcomes and the labor market of US households have been amply studied. Results are generally well established, for example that the gender gap has considerably narrowed, that the college premium has continued to rise and that female labor supply has increased a lot but is now stagnating. Claudia Goldin and Lawrence Katz have undertaken a new kind of analysis, studying the outcomes of Harvard graduates.

Why would we care? This is only a very small fraction of the population, not representative at all, and way out in the tail of the distribution. For one, they are over-represented among the decision-makers in the United States, and thus their personal experience influences their perception of what happens to others. Second, the education of the elite is typically heavily subsidized (privately or publicly), thus it is important to understand whether it is worth it.

This is an ongoing project, so not all questions have been answered yet. But the Harvard And Beyond Project website already offers a glimpse of some of the answers Goldin and Katz may have. Here are some results I found interesting.

1. The vast majority have pursued graduate degrees after their BA at Harvard. This proportion has increased for females and has remained stable for males. In the most recent cohort (1990 graduates), two thirds went for a graduate degree. In other words, for most, even a degree from the most prestigious program is not sufficient.
   
2. Harvard graduates earns a lot 15 years after graduation. For those with full time jobs, the median is at $112,500 for women and $187,500 for men. The gender gap is considerable, even after controlling for labor supply.
   
3. They marry later and have much fewer children than the general population.
   
4. They barely have any non-employment spells. Even among women who had children, over half never had an non-employment spell of more than six months. Penalties in earnings are severe for time off, especially among MBAs.
   

Thus if we have made much progress to reduce gender gaps and pregnancy penalties in the general population, the same cannot be said for the elite.

Why are retirees not buying annuities?

Retirees face a major uncertainty when they have to decide how much to spend from their assets: how long they are going to live. You do not want to spend too much in case you end up centenarian, but it would also be a waste to live thriftily when one ends up dying early. Luckily, the market provides a solution and has done so for centuries: annuities, which provide a constant stream of income as long as one is alive. This represents substantial welfare gains. Yet, very few people take advantage of this.

The literature has been puzzling over this for a while, see Jeffrey Brown. For one, public pension plans already provide some constant income. But this still leaves a lot of gain from annuities. Inflation risk is not a problem as annuity products are offered that are indexed, even even some that provide long-term care benefits. Bequest motives have been ruled out as well, as is risk-sharing within families.

Note also that in a lot of the life-cycle modeling with uncertain lifetimes, researchers use annuities to represent the dynamics of assets during retirement and death, because it is simple to model and because it is plain rational to buy annuities. The fact that this is not borne by facts is worrisome for some results coming out of this research.

In a recent NBER paper, Brown, Kling, Mullainathan and Wrobel may have found a solution to the annuity puzzle: it is all about how annuities are sold. If it is clearly labeled as a consumption insurance, then people are willing to buy it, as consumption risk is minimal. If it is sold as an investment, people realize it is very risky (as an investment): one may lose a lot if one dies early. This does not look like a rational reasoning, but the authors have conducted experiments that confirm that how annuities are framed matters crucially.

Then why would companies selling annuities frame it as an investment product? The authors conjecture that this is simply what they are used to. Maybe this research will show them how to change their marketing.

Economists and frustrating laymen

Reading this article in the Guardian on how some experts in biology just cannot seem to manage to make their point against laymen, I kept wondering whether economists have the same problem. Everyone is a self-proclaimed expert in Economics, both because of the confusion between policy goals (the realm of politicians) and policy means (the realm of economists) and because everyone is part of the "Economy."

Now, let me make sense of this long sentence. What are economists good at? Seeing the big picture, finding in which ways things interact in a general equilibrium sense, figuring out how incentives work. Thus, they are more detached from the nitty-gritty and can see the complexity of things, unintended consequences, and externalities. The layman is generally not equipped for this. This is why you need to be trained as an economist to figure out the consequences of policy, for example.

Yet, too often, economists do not manage to convince public and authorities about policy. Think about rent control, labor market over-regulation in Europe, gas taxation in North America, red tape in developing economies, or the overall lack of understanding about the role of market forces. When I speak with non-economists, I am regularly stunned by their reasoning and their unwillingness to understand the "true" Economics, mostly because it does not suit them, not because there would be a logical flaw. Quite frustrating, especially considering that these people then vote.

I am all for people to vote on policy goals. But when it comes to how those goals are to be reached, leave the methods out of the policy debate and let the experts take care of it. If you want to build a dam, you do not let people vote whether it is going to be made of concrete or earth, you let experts decide. And so it is when it needs to be determined whether a particular medical drug is safe or not. Or whether one should put fluoride in city water.

Set a goal for pollution reduction, then let experts take over. Set a goal for educational outcomes, and let experts figure out whether school vouchers are the way to go. Set a goal for dependence on Middle-East oil, then let experts get us there. Set a goal for school achievements, let experts get us there.

Enough for this rant. I promise to be more academic in the next posts.

Poetry in Economics

Do not worry, I am sticking to prose for this post. This about a paper with the same title by Hugo Mialon. He looks into how economists choose the titles of their papers and finds that 28% of titles include some form of rhetorical device. He also finds that this leads to four more citations for the paper.

While I agree that a title can tease someone into reading some work, there is a big step to generating a citation. Citations are not for titles, but for content (hopefully). Now, would the citer never have read the paper if it were for the title? Poetic titles often hide the true content of a paper, thus a case can be made for missed citations because of the title.

Then why does Mialon find four more citations for poetic titles? He focuses on article published in the top three journals. These papers are read anyway, poetic title or not. I think his result would be reversed when looking at working papers, where a poetic title is really hit or miss. His sample only has the hits.

The US housing aid bill: good or bad?

I cannot decide whether the new housing aid bill is good or bad. Here is why.

Given the current situation and the number of foreclosures, the forced moves of households and forced sales of homes generate huge transaction costs. They stem from disrupted lives, moving costs, loss of equity, empty houses depreciating, etc. We all know how costly it is to change homes. Imagine do this on short notice and multiply this by a large number. And on the banks' side, their loss of capital (as homes lose value and debt goes bad) hinders them to give credits that could be productive for the economy. And banks have to devote resources to manage the foreclosures. The housing aid bill avoids many of these costs by keeping people in their homes.

However, bailing out homeowners and banks comes at a significant reputation cost for the government. As many have pointed out, this bill rewards the absence of personal responsibility. Those who were responsible will have to ultimately foot the bill. As in many other examples, if the government shows that it is tough in such situations and does not act as a lender of last resort, such situations would never happen. Well, they still may happen if there is limited liability (but less frequently), hence the need for some regulation. But essentially a tough government leads to self-restraint.

In other words, we have a situation that can be improved, but we also face a time consistency problem. The ideal would be that somehow the government would be able to state that it would be helping this time, but never again. This administration being at the end of its term may help, but presidential candidates have shown no sign of a commitment to toughness.

The Economics of toilet seats

No, this is not about the industry that produce toilet seats, but rather about the habit for men to leave the toilet seat up. Hammad Siddiqi analyses this problem from a game theoretic point of view. Let me quote the entire abstract:

We model the toilet seat problem as a 2 player non-cooperative game. We find that the social norm of leaving the toilet seat down is inefficient. However, to the dismay of “mankind”, we also find that the social norm of leaving the seat down after use is a trembling-hand perfect equilibrium. Hence, sadly, this norm is not likely to go away.

It appears there is a literature on the subject: Jay Pil Choi argues that the selfish rule of leaving the toilet seat how one prefers it is the most efficient one. Any other rule (always up, or always down) inconveniences each member of one gender twice on each bathroom trip. With the selfish rule, one may not need to alter the seat, depending who came before, and never alters it after doing one's business. The always down rule is only efficient if the inconvenience to females is much larger than to males.

Hammad Siddiqi argues that a crucial element is missing from this analysis: conflict, or in other words game theory. One could do it like in a cooperative game, like Richard Harter. This assumes that members of a household minimize the joint cost of toilet seat moving. But what if they act selfishly, and possibly strategically?

Within a non-cooperative game, player can inflict penalties on others. In this case it would be females yelling at males for leaving the seat up. While leaving the seat always down remains inefficient, as in the two previous analyses, it is now a Nash equilibrium. There are often many such equilibria in a game, so this does not say much. But it is also trembling-hand robust, which means it is still the best strategy if the man (or the woman) sometimes forgets. Hammad Siddiqi concludes by asserting that if females realized that the always-down rule is inefficient, they would stop yelling and allow for a better strategy. There is hope.

It is all in the title

Publishing in Economics is also a lot about marketing your results. It is thus no surprise to me that US Economists dominate this profession: Americans are simply better at selling. One particular aspect of this marketing is the title of your paper. It pikes the interest of the potential reader.

One master of the art of title writing is Daniel Hamermesh. Here is a sample of his research:

Strike Three: Umpires' Demand for Discrimination

Cues for Coordination: Light, Longitude and Letterman

Changing Looks and Changing "Discrimination:" The Beauty of Economists

The Economics of Workalcoholism: We Should Not Have Worked on This Paper

The Value of Peripatetic Economists: A Sesqui-Difference Evaluation of Bob Gregory

Beauty in the Classroom: Professors' Pulchritude and Putative Pedagogical Productivity

Dress for Success -- Does Primping Pay?

Shirking or Productive Schmoozing: Wages and the Allocation of Time at Work

How ‘Grievous’ Was the Biblical Famine?

12 Million Salaried Workers Are Missing

When We Work

Of course, Hamermesh has a tack fro quirky research topics, which helps in finding original and intriguing titles. I notice also that his early work did not have much originality in the titles. I conclude this comes with experience, or with tenure.

Zimbabwe: how to beat hyperinflation

Zimbabwe is in a truly horrible situation, including crippling hyperinflation. I am not quoting any numbers here, they would be obsolete to quickly... Mugabe has royally mismanaged his country, but he is not eternal. So, once the economy is half-way decently managed, how could one get out of hyperinflation?

The obvious answer is of course: stop printing money. But this is not that easy. Without confidence in the currency, prices will continue to rise, at least for a while. Government expenses still need to be met in the meanwhile, in particular civil servants, and there is no meaningful way to raise taxes. But even with a sound fiscal policy, there is no guarantee that hyperinflation can be beaten. For example, Beatrix Paal shows that a policy of stabilizing inflation though the management of public debt is essentially indeterminate and can lead to hyper-inflation, thus being ineffective. Thus, the critical point is to restore confidence in monetary policy while continuing to pay for government services.

Makinen and Woodward show that Taiwan managed to beat hyperinflation by guaranteeing high real interest rates on bank accounts. This seems like classic anti-inflationary monetary policy to the extreme.

I think the best solution is to delegate monetary policy to an entity outside of the country. This is what Cooper and Kempf advocate: dollarization or a currency board. Dollarization means abandoning local currency in favor of another one. But if one ultimately wants some control over monetary policy, this does not sound like a good policy.

In a currency board, the exchange rate with a base currency is set by law, thus difficult to change unlike a fixed exchange rate regime. In addition, every bit of currency in circulation is backed by the base currency (or titles in said currency): anybody can exchange at the predetermined rate at the central bank. Such currency boards have been very successful in creating confidence in the local currency in various countries from the former Eastern European Block, where the IMF helped building up the necessary foreign reserves. So far, the only currency board that failed was the Argentinean one, due to unsustainable local fiscal policy.

A currency board has worked superbly to end hyperinflation in Bulgaria, as shown by Stefka Slavova. But as the Argentinian example shows, and as Thomas Sargent argues in a celebrated book, sound fiscal policy is still needed from the start, but it is not sufficient, as also Gustavo Franco showed.

The IMF mission to the US: embarrassment or normal procedure?

One of the roles of the IMF is to make assessments of economic policies in member countries and forcing them to adopt sounder ones. The important word here is "forcing." Many governments are in fact grateful for this, as it allows to enforce good, but unpopular policy using the IMF as a scapegoat.

In principle, any member country could be subject to such scrutiny. Unfortunately, there is considerable politicking in the IMF, and in particular rich countries manage to impose upon others prescriptions they would adopt themselves. They can get away with it due to current structure of the IMF. We reported before on the need for this structure to be reformed.

In turns out the US will be scrutinized soon within a Financial Sector Assessment Program (FSAP), i.e., a complete analysis of the financial sector. Market participants and government agencies will be required to hand over confidential documents. This is no different than what is done elsewhere, but the uproar is certain to appear.

One could view this as a sign that finally rich economies are coming under the same scrutiny as the poorer ones. Not quite. Indeed, this mission had been on the radar for a long time, but the Bush Administration vehemently opposed it for seven years, but finally gave in on the condition that the report be issued after the handover to the next administration. By then, everyone in charge will be out of office, but one: Ben Bernanke.

Speculators are not the problem

The US Congress has finally found the scapegoat for high oil prices. While most of the increase is due to the fall of the US dollar that will eventually rectify itself, it found speculators to be guilty and intends to rein them in. This is nonsense.

Speculators are regularly vilified because they manage to make money without being apparently productive. Yet, they provide some very important functions on the market: they help hedging against risks, and more importantly, they help prices being informative of true economic conditions. Through their arbitraging, they make sure that goods are properly priced. For example, if an under-supply of oil is expected, they make sure that the price of oil reflects this. This allows producers and consumers to adjust to conditions. If prices would not reflect markets conditions, rationing could appear.

Yet it seems Congress wants exactly that: prices that do not reflect economic conditions. The current proposal is to inhibit the ability of speculators to arbitrage by preventing them to resell (which is at the core of arbitraging) and to deal with foreign markets.

There is no question that whenever markets are manipulated, intervention is necessary. How do you define manipulations? Albert Kyle and S. Viswanathan define a two-pronged test: Price manipulation needs to simultaneously undermine both pricing accuracy and market liquidity. In other words, their is manipulation if prices do not provide signals about economic conditions while there is evidence that someone is preventing trades from happening. Prices could be poor signals in liquid markets, but their is nothing one can do and nobody is benefiting from it. Prices can be accurate in illiquid markets, and that is not a problem. But both happening at the same time is a sign of manipulation.

Are oil prices currently manipulated? Given the size of the market, this is unlikely. But it has happened before, for example when the Hunt brothers manipulated the silver bullion market in 1979-80. At that time, they severely curtailed the liquidity of the market by hoarding. That does not seem to be the case with oil today. Oil markets are very liquid, and prices do reflect a real scarcity in addition to risk.

Note that what Congress proposes would be considered price manipulation, as transactions are prohibited (which lowers market liquidity) and prices likely would not reflect economic conditions (which undermines pricing accuracy). In other words, the proposal would make things worse... But it is responding to the call of doing something, and this is what counts in politics, right?

The Economics of energy subsitution

The increase in oil prices allows nicely to highlight the mechanics of substitution. The increase in the price of most goods lead to a decrease in its use, while increasing the demand for its substitutes. This leads to an increase in the price in the other goods. We have seen this in the past month nicely with increases in electricity and food prices, although these are not pure substitution effects (oil is at least partly an input).

Another substitution effect come form the use of goods that were not used before. In the case of energy, using alternatives like solar energy or windmills becomes more economical, thus creating goods that were not in demand before. But again, this is not a pure substitution effect, because these alternative energy sources have been pushed for other reasons as well, such as pollution reduction.

For automobiles, the rise of hybrid cars is a substitution effect, although they still use some gas. What about a car that does not use energy from oil at all? Enter the AirCar, which simply runs on compressed air. The concept is ten years old, but was not economical until now (except for some cars running in Spain). Tata Motors, the major Indian car manufacturer now announced it will start producing a car based on this concept in August 2008. The MiniCAT will have a range of 300km for a maximum speed of 105km/h, the refill will come to $2.00 at a station, and an emergency compressor can be plugged into a socket to refill as well.

Note that this car does not use the air pressure per se, but rather the thermodynamic effect when you change the pressure and the volume of the air. The emissions are thus only very cold air, which can be used for air conditioning...