Brief reflections on the portfolio risk the impact of withdrawals

Portfolios rarely live in perpetuity and even when they do there are numerous occasions in their lives when cash flows have an impact on their structure and performance.

We usually measure portfolio risk by determining the volatility or standard deviation of its returns. However, in the context of a well-diversified, balanced portfolio, the major short- and medium-term risks are associated with portfolio withdrawals (or distributions). In the absence of distributions, the beneficial impact of reinvested cash flows (from dividends and coupons) and compounding are enough to overcome, given sufficient time, almost any conceivable market setback.

The problem with withdrawals is their amount and timing: if, for example, distributions take place in difficult market circumstances and if their amount is excessive in relationship to the size of the portfolio, then the problem is almost impossible to manage without depleting the real value of the portfolio itself. Various analyses that show how, with time, certain random scenarios impact the future evolution of the value of a portfolio (with or without distributions) are seductively interesting, but they hide a truth: the scenarios are based on averages, while your portfolio is a single occurrence. If you are hit by three bad years consecutively at the beginning of the investment period, no amount of previously identified average scenarios will help you.

What are the best circumstances under which to handle distributions? The most important consideration is flexibility: the more flexibility one has in determining portfolio distributions (in size and timing), the less impact these will have on the investment strategy of the portfolio itself. Flexibility in amount and timing is the best of all worlds, as amounts can be determined in the context of expected and realized portfolio returns and timing can be maneuvered around market conditions. Flexibility in amount is a second best, followed by flexibility in timing. The worst case is where there is no flexibility in the timing and amount of distributions, and where these are close to or above the conservatively estimated return potential of the portfolio.

There is then a direct relationship between investment strategy (or appropriate portfolio risk) and flexibility in managing portfolio distributions: the more inflexible the administering of withdrawals is, the more the portfolio structure and strategy should approach a fully immunized portfolio (essentially matching cash outflows with both income and capital). To think that active management alternatives are a solution is pure speculation.

In conclusion, if we have flexibility in amount and timing of portfolio distributions (from Credit Suisse, for example) we also have freedom of choice of investment strategy. As the flexibility diminishes and the obligation of making the payments becomes gradually more rigid, either in amount or timing or both, then our ability to invest and take advantage of what the markets can offer us diminishes as well. There is unfortunately no realistic alternative that I know of to this analysis; no amount of speculation or projections about returns can counter the lack of flexibility in portfolio distributions.