Never Spend Principal?

Never spend principal.  Perhaps no other precept of investing has been passed from generation to generation so solemnly, and for good reason.  Restricting spending to current income (primarily interest and dividends) can avoid a lot of trouble in an equity-oriented account, particularly in down markets.  However, this rule was much easier for prior generations to follow.  From 1926 through 1990, the average annual dividend yield on the S&P 500 Index was 4.7 percent.  The S&P 500 Index currently yields only 1.8 percent.  Interest rates have also fallen below their long-term averages, although with the recent rise in interest rates, the decline in bond yields is less dramatic.

As yields have dropped, total return approaches to investing and spending have become increasingly popular.  The phrase "total return" in its narrowest sense means the combination of the two elements of investment return: appreciation (or depreciation) and current income.  "Total return" is also used as shorthand for approaches to spending that are not completely reliant on current income.  For example, many non-profit institutions have a spending policy of using 3 to 5 percent of the value of their endowments to support their operations each year.  Such policies are said to be total return approaches because they allow investment strategy to be driven by the quest to maximize total return, regardless of the level of current income generated.  While not without risks, we think total return policies can make sense if they are based on a realistic view of investment returns and spending goals (inclusive of expenses).

An Historical Perspective

For many years, charting a sustainable spending policy was relatively uncomplicated.  The 1980s and 1990s delivered markets characterized by both high yields, which reduced the need to spend principal, and better than average appreciation, which replaced any reasonable principal withdrawals.  For example, in 1982--when the yield on the 10-year Treasury was around 14 percent, total return for the Lehman Brothers Aggregate Bond Index was above 30 percent, and the S&P returned over 20 percent--it scarcely mattered if one withdrew 10 percent of one's portfolio in a year.  However, the current decade has brought an entirely different experience to investors; yields and returns have thus far been much more modest, and have sometimes even been negative.

Meet Mr. and Mrs. Brown

Let's take a closer look at this market scenario through the experiences of John and Sarah Brown, our hypothetical investors.  They were very fortunate when they sold their business in late 1990 and decided to retire.  Through accumulated savings and the after-tax proceeds from the sale, they had a liquid portfolio of $10 million.  Both aged 55, they needed the portfolio to support their lifestyle for many years to come.  They invested 60 percent of their portfolio in equities (broadly diversified across domestic and international stocks) and 40 percent in bonds.  Using a "back of the envelope" approach, John and Sarah assumed that with this healthy allocation to equities they could support a generous lifestyle over the long-term.  They understood and relied on the fact that equities had double-digit returns on average in the prior decade.  They decided to take a total return approach and, without the benefit of prudent advice, planned to withdraw $1 million--grown with an assumed inflation rate of 3 percent--from the portfolio each year (10 percent of its initial value) for their spending needs as well as taxes and fees.  With the benefit of hindsight, we can use actual market returns over the last 15 years to see if the Browns achieved their goals. 

Using historical index returns, figure 1 on the next page charts the health of their portfolio from 1991 through 2005.  At the end of 2000, the value of the portfolio would have exceeded $13 million.  In other words, even with significant value removed from the portfolio (over $11 million cumulatively) to support the couple's lifestyle, the portfolio would still have grown by more than 30 percent, thereby keeping about even with inflation.  Unfortunately, though, as the Browns continued with the same level of withdrawal for the next five years (2001 through 2005, a more difficult period for stocks), their fortunes changed dramatically.  Instead of modest growth, the portfolio experienced a significant decline, leaving the Browns with less than $8 million.  After inflation, their portfolio came to be worth about half of the initial $10 million.  Even the strong returns in 2003 could not provide much of a rebound--by that time, the inflation-adjusted withdrawals had swelled to greater than 15 percent of the portfolio value.  It would not have been a wise strategy for the Browns to spend such a significant portion of their assets each year.  Their implicit assumption--and it turned out to be wrong--was that the portfolio would perform at a consistent rate above 13 percent a year (10 percent withdrawal plus 3 percent inflation). 

How would we have advised the Browns? Our experience and our research has shown that portfolios, no matter what the asset allocation, would be overwhelmed over the long term by a rate of consumption as high as the Browns'. We feel that, in general, a more reasonable annual spending rate is in the 3 to 5 percent range (inclusive of taxes and fees) depending on long-term goals, risk tolerance and asset allocation. If an investor's goals for the portfolio include keeping pace with inflation or investing conservatively, then annual spending should be kept to the low end of our range, closer to 3 percent. On the other hand, if the goal is to maintain nominal wealth, to slow the decline in the portfolio (due to consumption) or to invest more aggressively, then a 5 percent spending could be appropriate.  Reaching the right conclusion for each investor requires a sound planning process.  This process, as we follow it, is detailed below.

Step One: Be Reasonable with Assumptions

As you can see from the example above, relying on above-average performance from the markets was not helpful to the Browns.  When planning, the first step is to adopt a set of reasonable assumptions, beginning with the long-term historical returns on equities and fixed income as well as a prospective look at what could happen in the future.  Since the end of 1925, U.S. large-cap stocks (as represented by the S&P 500 Index) have returned slightly more than 10 percent annually, but we believe 9 percent is a more conservative assumption for the next decade or so.  A municipal bond portfolio currently yields about 3.7 percent.  Thus, the Browns, taxable investors with a balanced portfolio of 60 percent equities and 40 percent municipal bonds, might reasonably anticipate long-term returns of about 7 percent, quite different from their initial expectation of at least 13 percent.  Of course a number of factors, including good active portfolio management and the inclusion of small-cap stocks, international stocks, real estate and alternative investments, can boost long-term results; for planning purposes, though, 7 percent is a reasonable expected return.

It is also important to remember that the Browns do not get to keep or spend all of those returns.  We estimate that the taxes on the equity portion of this portfolio would reduce the returns by about one half of a percent annually.  Investment advisory fees and other professional fees also need to be taken into consideration; consequently, the Browns should not plan to "pocket" more than 5.5 percent to 6 percent annually, assuming that the return patterns were constant.  Knowing that return patterns are never constant, an understanding of variability of returns in the context of their plan is critical.

Step Two: Evaluate Goals and Risk Tolerance Together

Once the Browns have arrived at reasonable return and spending assumptions, they need to match them with their long-term objectives and their tolerance for volatility in the portfolio.  If they want their portfolio to keep up with inflation, which has historically averaged about 3 percent, spending more than 3 percent of the portfolio may be a problem if they maintained the 60/40 asset allocation discussed above.  However, keeping pace with inflation is not a primary objective of many investors.  Most individuals build wealth during their earning years and draw down on it during retirement.  In either case, the anticipated withdrawals and the risk level of the asset allocation must be measured against the investors' goals.

The simple analysis above assumes that the Browns are willing to have 60 percent of the portfolio invested in equities.  For some, this level of equity investment can generate uncomfortable levels of near-term volatility; should this be the case, the allocation to equities can be lowered in favor of fixed income or other less volatile asset classes.  Based on these return assumptions, if the Browns allocated 40 percent to large-cap equities and 60 percent to fixed income, the expected total return would then be about 5.5 percent.  As long-term return expectations come down, so must consumption.  With a $10 million portfolio, choosing a $300,000 lifestyle versus a $600,000 lifestyle with a conservatively allocated portfolio will meaningfully impact the Browns' financial security over the long-term. 

There are two important differences in the asset allocations mentioned above, the return assumption and the expected volatility in the portfolio.  First, a portfolio of 60 percent stocks should generate higher long-term returns than one with 60 percent fixed income instruments.  Second, while the average return over a long period of time will generally be expected to be greater with a higher allocation to equities, the years in which there are losses could be more frequent and more severe than for a portfolio with less equity exposure.  In other words, it will almost always require a stronger stomach to deal with the fluctuations in the equity markets when the portfolio is oriented toward stocks.  The Browns need to be realistic about their tolerance for risk so that they do not reach for a higher return to meet their lifestyle expenses and then, down the line, lose their nerve and sell out in a down market and thereby lock in losses.

Step Three: Use Market Simulation Tools As A Guide

To lend some guidance to this process (and to families like the Browns), Brown Advisory employs a market simulation tool that matches hypothetical portfolios and goals over thousands of possible capital market scenarios.  The result is a forward-looking analysis (based both on historical data and our prospective view of the markets) that allows an investor to make better-informed decisions about risk, volatility, and, in this context, appropriate consumption rates.  While the best plan is perfectly tailored to the individual facts and circumstances, the example of our hypothetical Browns is instructive.

Looking at a portfolio with an initial value of $10 million (the level from which the Browns began their journey), we tested annual withdrawal strategies ranging from 3 percent to 8 percent of the initial portfolio value.  These withdrawals are assumed to cover lifestyle expenses, taxes, and fees.  We assumed the lifestyle expenses were fixed as a dollar amount and grown with inflation over time and that the taxes and fees were recalculated annually as a percentage of the asset base.  We also tested the impact of withdrawals across a broad spectrum of asset allocations, from conservative (20 percent globally diversified equities, 80 percent fixed income) to aggressive (80 percent globally diversified equities, 20 percent fixed income) with more moderate mixes in between. 

Testing for Spending Limits

Figures 2 and 3 show values over time of a portfolio invested 60 percent in equities and 40 percent in fixed income with a wide range of spending strategies.  Our simulation process produces results for 10,000 random market scenarios--the good, the bad and the ugly.  Figure 2 is based on the returns from average markets, whereas figure 3 shows the same analysis for bad markets--the 90th percentile of the 10,000 scenarios generated. 

Even in average markets, the results are eye-opening.  An initial 3 percent rate of withdrawal allows a moderately allocated portfolio to continue to grow with a high likelihood of keeping pace with inflation.  A 5 percent withdrawal allows for moderate nominal growth with the portfolio losing ground on an inflation-adjusted basis.  An 8 percent initial withdrawal, on the other hand, overwhelms the portfolio over the long-term--this is an example of a plan that simply isn't prudent when one considers the client's circumstances.

Determining An Asset Allocation

Figure 4 shows the impact of asset allocation for a 4 percent initial withdrawal.  With a 4 percent   withdrawal pattern in average market conditions, it takes an allocation of at least 60 percent in equities to grow at a rate greater than inflation. Being too conservative with the investments can jeopardize long-term withdrawal goals and cause a meaningful decline in the portfolio in later years.  It is clear that asset allocation plays a large roll in the overall planning and underscores the need to look at anticipated withdrawals and portfolio construction together when plotting a course.

Conclusions

In developing the right long-term strategy, it is important to recognize that while history is an invaluable guide, we do not know the pattern or magnitude of future returns in the capital markets.  This makes prudent planning all the more critical.  Understanding the financial goals, setting reasonable assumptions and judging the tolerance for risk in the portfolio are all essential steps.  Once goals are established and risks evaluated, we can use the tools available to assist investors in making decisions.  Each investor's situation is different, and individualized analysis is critical, but our experience tells us that there are a few basic tenets of a successful spending plan:

For taxable investors whose goal is to have their portfolio's purchasing power keep pace with inflation, withdrawals for living expenses, taxes, and fees probably should not exceed 3 percent of the initial value of the portfolio (plus subsequent inflation adjustments).

For retirees, withdrawals for living expenses of around 5 percent of the initial value of the portfolio will often produce satisfactory results, but probably will not preserve the full purchasing power of a balanced portfolio in the long run.

If possible, it is best to fix withdrawals for living expenses as a percentage of the initial value of the portfolio and factor in an annual adjustment for inflation.  Recalculating withdrawals on a percentage basis when returns have been good can deprive the portfolio of the opportunity to compound excess returns.  Moreover, experience suggests that clients usually do not adjust their spending downward when returns have been poor.  This approach will probably provide opportunities for increases to withdrawals over time, but we think such increases should come as part of a prudent planning process, not the mechanical operation of a percentage payout.

While the long-term returns of an equity-oriented portfolio will likely be higher than one dominated by fixed income, asset allocation decisions need to be made with an understanding of the anticipated consumption rate and the investor's tolerance for volatility in the portfolio in the near- and long-term.