Spending strategies for retiring in a down market

By Joana Rocha, Investment Strategy Analyst, Vanguard, Europe

Market swings are a normal part of the investing cycle. But they can pose particular risks for those nearing or in retirement who plan on making regular withdrawals from their savings.

For these clients, the double-whammy impact of beginning their withdrawals at a time when their portfolios are in decline makes it doubly difficult for them to fully recover their losses when market conditions improve—thus potentially setting off a string of further shortfalls that could lead to their savings running out too soon.

This scenario—known as sequence-of-return risk, or sequence risk— may have an enduring impact on their long-term retirement outcomes.

To help illustrate the impact of sequence risk, let's look at an historical example from the 1970s that compares two similar investors who retired one year apart: the first investor began retirement in 1973 at the start of a severe bear market, and the second entered retirement a year later in 1974, at the tail end of the downturn¹. Despite a difference of just one year in the timing of their retirements, the two investors experienced dramatically different outcomes, with the 1973 retiree running out of money in 23 years and the 1974 retiree enjoying a healthy portfolio balance that provided them with a steady income for the entire 35 years of their retirement.

Sequence-of-return risk: Same generation, one-year return difference in retirement wealth over time

Notes: This figure assumes two hypothetical investors retiring at the beginning of the calendar year with $500,000 portfolios invested 50/50 in stocks and bonds. Both withdrew a fixed amount of $25,000 per year (inflation-adjusted). This hypothetical illustration does not represent the returns on any particular investment and the rate is not guaranteed.
Sources: Vanguard calculations, based on data from Morningstar, Inc.; the Federal Reserve Bank of St. Louis; and the Kenneth R. French Data Library. All data are monthly and for the period January 1973 to December 2008. All income is assumed to be reinvested, and no fees or taxes are included. Calculations are in USD.

Both retirees entered retirement with $500,000 portfolios invested in 50/50 stocks and bonds and planned to withdraw a fixed amount of $25,000 a year (inflation-adjusted). Their only difference was the 1973 retiree began withdrawing at the start of a sustained market decline (January 1973) whereas the 1974 retiree began their withdrawals one year on, when the stock market was still moving lower but closer to its trough (September 1974).

The longer the downturn, the harder it becomes for the portfolio to recover in between withdrawals—the results of which can be devastating. In the case of the 1973 retiree, their portfolio never fully recovered from their early withdrawals, leading to the full depletion of their savings two decades later; by contrast, the 1974 retiree’s portfolio maintained a balance of around $300,000 for most of their 35 years of retirement, finishing with a bequest worth around 25% of their pre-retirement portfolio.

Many other factors—frequency and size of withdrawals, other sources of income, longevity—can influence the impact of a choppy market on a retired client’s portfolio longevity, but the 1970s example highlights the impact that fixed withdrawals can have on an already weakened portfolio and the increased risk they can pose to a retiree’s financial well-being over the long run.

Developing a spending strategy for retirement

For advisers, helping clients manage their retirement spending can be one of the most valuable ways of protecting their portfolios in retirement—not to mention the emotional value of helping them feel more in control as they transition into the decumulation stage of their lives.

Traditionally, the most popular retirement spending strategies have been the “dollar-plus-inflation” approach—where the retiree sets an initial spending sum that increases by inflation in each subsequent year, as used by the retirees in our 1973/74 example—and the “percentage of portfolio” approach, where spending is based on the retiree’s portfolio value that fluctuates with market conditions.

While both strategies can help clients budget their spending when markets are healthy, neither is ideal in a downturn. The “dollar-plus-inflation” approach offers a higher degree of certainty around withdrawals, but also comes with higher risk of the underlying portfolio being depleted. On the other hand, the “percentage of portfolio” rule reduces sequence risk, but assumes that retirees can quickly adapt their spending levels based on portfolio performance—which may be unrealistic in down years if their withdrawals are insufficient for covering expenses.

To help address these needs, Vanguard developed a hybrid option called “dynamic spending”, that allows for fluctuations in annual withdrawals based on portfolio performance while providing a level of income stability to support the client’s individual spending needs. The strategy works by modestly adjusting the withdrawal amount in response to portfolio performance each year, subject to a ‘ceiling’ of 5% when markets fare well and a ‘floor’ of -2.5% when they don’t.

Dynamic spending in practice

While dynamic spending can extend the life of a retired client’s savings, advisers can play a pivotal role in helping tailor the strategy to the needs of their individual clients.

To illustrate how dynamic spending works, let’s draw on an example: 

Mrs Smith is retiring with a $500,000 portfolio and a target withdrawal rate of 5%—providing her with an initial income of $25,000².

At the start of Year 1, Mrs Smith takes her initial income of $25,000—leaving her portfolio worth $475,000.

Mrs Smith’s target income for Year 2 is $25,500 ($25,000 + 2% inflation). This can fluctuate as high as the 5% ceiling ($26,775, or 5% above $25,500) and as low as the -2.5% floor ($24,862, or 2.5% below $25,500).

Spending scenarios

Unlike a fixed withdrawal strategy, Mrs Smith’s actual spending budget in Year 2 will be determined by her portfolio performance during Year 1.

Let’s consider three different portfolio performance scenarios:

Scenario 1: Portfolio increases 8%

Mrs Smith’s portfolio increases 8%, from $475,000 to $513,000. Based on a 5% target withdrawal rate, the target income would be $25,650. Because this falls within the dynamic spending band, Mrs Smith should withdraw the full $25,650.

 

Scenario 2: Portfolio decreases 4%

Mrs Smith’s portfolio decreases 4%, from $475,000 to $456,000. Based on a 5% target withdrawal rate, the target income would be $22,800. Yet this falls below the “floor” guardrail sum of $24,862. Mrs Smith should therefore withdraw $24,862—more than she would have done if she had factored in the full market performance.  

 

Scenario 3: Portfolio increases 20%

Source: Vanguard

Smith’s portfolio increases 20%, from $475,000 to $570,000. Based on a 5% target withdrawal rate, the target income should be $28,500. Yet this is above the “ceiling” guardrail sum of $26,775. Mrs Smith should therefore only withdraw the ceiling sum of $26,775—less than if she had factored in the full market performance.

How effective is dynamic spending?

Our research shows that dynamic spending in a drawdown can greatly improve retirement outcomes compared to fixed-sum withdrawals under the “dollar-plus-inflation” approach³. By adjusting withdrawals to partially match portfolio performance throughout retirement, a retiree can greatly reduce their risk of running out of money over the long term, while also increasing the level of their wealth remaining at the end⁴.

To better understand the impact of dynamic spending in a recession, researchers re-modeled the retiree portfolios from the 1973/1974 scenario using a dynamic spending strategy and compared the results against the original “dollar-plus-inflation” approach. As the chart below shows, both retirees greatly improved their retirement outcomes using dynamic spending; critically, the 1973 retiree avoided the premature portfolio depletion that occurred with fixed withdrawals.

Past performance is not a reliable indicator of future results.
Notes: A hypothetical investor retiring in the beginning of the calendar year with $500,000 invested in a portfolio invested 50/50 in US stocks and US bonds, rebalanced monthly, and fixed withdrawals of $25,000 per year (inflation-adjusted). All data are monthly and for the period January 1973 to December 2010. All income is assumed to be reinvested, and no fees or taxes are included. The oil crisis ran from October 1973 – March 1974. Black Monday was 19 October 1987.  The Tech bubble burst in 2000. The ‘GFC’ refers to the ‘Global Financial Crisis’, which occurred from 2007 to 2008.

Sources: Vanguard calculations using data from the Kenneth R. French Data Library; intermediate-term US government bond returns are from Morningstar’s Yearbook on Stocks, Bonds, Bills, and Inflation; and US inflation data is from the CPI Index for All Urban Consumers. All calculations are in USD.

The value of retirement advice

During times of uncertainty, dynamic spending strategies can help retired clients weather the effects of market shocks on their portfolios while helping them preserve a reasonable level of income to meet their spending goals.

Regardless of the spending strategy chosen, the complex interplay of fluctuating market conditions and the evolving needs of individual clients as they progress through retirement underscores the need for, and the value of, skillful guidance from a knowledgeable adviser on an ongoing basis.

Find out more

To find out more about how you can apply dynamic spending with your clients, please contact your Vanguard relationship manager or get in touch here.

 

¹ The 1973/74 bear market lasted 21 months, beginning in January 1973 (peak month), and bottoming in October 1974 (trough month), with a full peak-to-trough decline of 46.4%. Source: Vanguard calculations, using data from Kenneth R French Data Library.

² Unless otherwise stated, all dynamic spending calculations use standard income guardrails of a 5% ceiling and -2.5% floor. The long-term expected inflation rate is assumed to be 2%.

^3,4 Source: Vanguard calculations, based on data from Morningstar, Inc., the Federal Reserve Bank of St. Louis, and the Kenneth R. French Data Library. For more details, please see “Safeguarding retirement in a bear market”, Khang, PhD, Kevin I; Andrew S. Clarke, June 2020.