Friday, October 28, 2011

Are retirement goals fixed or flexible?

One of my early forays into research about personal financial planning was, “Lifecycle Funds and Wealth Accumulation for Retirement: Evidence for a More Conservative Asset Allocation as Retirement Approaches” from the Spring 2010 issue of Financial Services Review.  The issue I was responding to was research by Harold Schleef and Robert Eisinger which argued that retirement savers are better off by maintaining more aggressive stock allocations near retirement, rather than decreasing their stock allocation as is done with retirement target date funds. I thought that their approach for reaching this conclusion didn’t sound particularly compelling. They defined retirement success only in terms of whether or not an individual reached their retirement wealth target by their target retirement date. Whichever asset allocation experienced the highest probability of reaching the wealth target was “optimal” in their framework.

Stocks tend to offer both higher average returns and higher volatilities over long periods of time. Because of the higher average returns, stocks will tend to maximize the probability of reaching a wealth target. This explains the findings by Schleef and Eisinger.  But what I think must also be considered was the fact that the distribution of wealth provided by higher stock allocations is also wider, and this includes the possibility of experiencing much lower wealth accumulations as well. 

I thought it was important to consider the whole distribution of outcomes. Here is what I wrote in my article:

Our findings will tend to provide some support for the use of target date funds.  We argue that it is important to focus on more than just meeting a particular goal for retirement.  The simulation approaches used by studies such as Schleef and Eisinger (2007) and Basu and Drew (2009) provide an entire distribution of wealth outcomes, and researchers have an opportunity to take advantage of all this information.  The basic issue is this: For someone whose goal is to maximize their mean or median wealth accumulations at their retirement date, then it is clear from historical trends that the best chance for success is to maintain a high equity allocation near retirement, in contrast with the general philosophical approach of target-date funds.  A risk averse individual, however, may have a different goal, such as minimizing the risk of suffering from extreme hardships in retirement.

Since writing that article, I’ve added more reasons why retirement savings shouldn’t be so focused on meeting a wealth accumulation target [my “save savings rates” paper argues that savers should focus on their savings rates and not worry about accumulated wealth or required withdrawal rates, and my “getting on track for a sustainable retirement paper” argues that it is quite difficult in practice to know whether you are on track to meeting a wealth accumulation target anyway].

But for today, the issue I want to focus on is how there is nothing magical about such a wealth target.  You don’t automatically succeed because you have one dollar more, or fail because you have one dollar less.  Goals are fungible, and people will adapt to having either more or less. To determine an asset allocation strategy for retirement savers, it is important to evaluate the whole range of potential outcomes and choose the distribution of outcomes that will maximize your expected lifetime satisfaction. A wealth target is not an all-or-nothing affair.

Now, I’d also like to argue that this extends as well to retirement spending. I don’t think retirement success or failure depends on whether or not one meets some particular goal.  I’m inspired to write this by a blog entry which Michael Kitces wrote last December.  He wrote:

The problem is that, if you're focusing on what it takes to achieve a client's goals, just insuring 50% of a goal seems to be a remarkably inadequate solution, if someone is really concerned about making sure that the goals are achieved. After all, if a couple's goal is to spend $60,000/year for the rest of their lives, then if "the bad stuff" happens and they must rely on the $30,000 (50%) that was insured, then haven't they still catastrophically failed to achieve their goals? The good news is that they won't be destitute without income. But if the purpose of the plan was to achieve the goal - the goal has still been failed!

To answer the question Michael asks, I think the answer is “not necessarily.” Everyone is different, and for some people this could represent a catastrophic failure. But more generally, I want to argue against the general acceptance of this sort of all-or-nothing retirement spending goal as being the conventional wisdom for retirees.

As Bob Curtis explains in his article, “Monte Carlo Mania” from Harold Evensky and Deena Katz’s book, Retirement Income Redesigned: Master Plans for Distribution, retirees do not have just one spending goal. They instead have different goals, each with a different priority. There may be some absolute spending floor which is seen as essential, and then some other discretionary spending on top of that, followed by a desire to own a nicer car and to enjoy one trip abroad each year, and then there is the possibility to provide some gifts to charity and relatives, and if that is taken care of than some additional spending toward enjoying more nice restaurants is a final possibility.

So to define what an all-or-nothing goal is, where would we draw the line for this person?  Does it include everything I listed?  Because that may greatly intensify the probability that portfolio wealth will be exhausted before death.  So which goals should we cut in order to provide sufficient chances of being able to adequately cover the most important goals? The answer depends on how important each goal is, or how much utility each goal provides. When the goals are listed in the order of their priority, each successive goal must be providing less life satisfaction per dollar than the previous goal. That is why its priority is lower. Financial planners can help their clients to think about how much satisfaction each goal provides, and then to focus on the tradeoffs between satisfying more goals in early retirement which increases the chances of having to cut back on other goals later in retirement.

Planning for retirement is complex, unfortunately. We don’t know how long we will live and we don’t know what will happen with financial markets in the coming years. So we have to do the best we can to balance some complex tradeoffs.  I am strongly in favor of making this process as simple as possible, but I really fear that thinking in terms of one all-or-nothing retirement spending goal is too simplified and could lead retirees to make mistakes which don’t provide them with the most expected lifetime satisfaction from their hard work and savings.