Michael E. Kitces, a partner and director of research at Pinnacle Advisory Group, a private wealth management firm, tackled the topic of safe withdrawal rates at last week’s CFA Institute Wealth Management 2013 conference in Boston. Kitces, who publishes the e-newsletter The Kitces Report and the blog Nerd’s Eye View, also provided an overview of current research on optimal withdrawal rates along with an analysis of historical safe withdrawal rates recorded in varying return environments.
In his presentation, Kitces noted that there are two fundamental client questions: How much can I safely spend from this portfolio without needing to worry about the markets? And, if I want to spend X, how much money do I need in the account to safely retire?
In a standard spreadsheet for planning a client’s retirement, you will have a ledger chart that looks something like this: starting balance + growth – withdrawals = end-of-year balance. If the numbers don’t yield a positive balance, you need to move one of the levers that will make retirement work for the client. There are really only four: save more, spend less, retire later, or die sooner.
Kitces explained that the sequences of returns matter (a lot). Disparities in the early years have a magnified effect over time, and the extent of volatility also matters.
Some of the challenges of safe withdrawal rates:
Given the impact of volatility, how much of a “safety margin” is necessary?
Given the historical returns of the markets, how high of a withdrawal rate would have survived any historical market scenario?
What is the optimal portfolio allocation to survive the volatility?
The key take-away from the discussion: “Safe withdrawal rates are built around ‘worst-case’ scenarios, they are not built around average returns. It has nothing to do with average returns,” Kitces said.
While the low-return environment puts safe withdrawal rates at risk, it is not necessarily fatal to success.
Kitces has found there is a dramatic misunderstaning of what safe withdrawal rates really assume. In reality, the current return environment does not substantively violate the withdrawal rate assumptions — it actually matches the assumptions. (Safe withdrawal rates never assumed average returns; they’re built for 0%-real-return environments.)
For more on this topic, see his related blog posts: What Returns Are Safe Withdrawal Rates REALLY Based Upon? (August 2012) and Safe Withdrawal Rates In Today’s Low Yield Environment – Walking On The Edge Of A Cliff? (February 2013).