I haven’t talked about safe withdrawal rates (SWR) in a while. For earlier discussions on what the SWR is on how to use it in retirement see the series of posts here. The SWR basically determines what is the maximum a retiree can withdraw from their portfolio every year and make sure the portfolio lasts through their entire retirement. One of the harder parts of implementing the SWR is making increases in withdrawals at times when the value of the portfolio is heading down. It often takes immense emotional fortitude to trust in history and increase spending in a given year a portfolio just took a 20%+ dive, as happens quite often in markets. After all, the past is no guarantee of future results. Which makes you wonder if there is an early warning indicator for a retirement portfolio that can help a retiree determine if adjustments to spending are necessary? There is such an indicator and its called the current withdrawal rate (CWR).

The SWR represents the percentage of a retirement portfolio withdrawn in the first year of retirement which is then adjusted annually for inflation. The current withdrawal rate would be the dollars withdrawn in a given year divided by the current portfolio value. For example, assume the SWR for a retiree is 4% and the initial portfolio value is $1M. Year 1 withdrawals would be $40K. Also, assume in year 2 inflation is 5% and the portfolio increases by 5% in that year. Year 2 withdrawals are $42K and thus the current withdrawal rate (CWR) in year 2 stays at the same 4% ($42K/$1.05M). All good. Alternatively, if inflation is 5% and the portfolio declines 10% in year 2, the the CWR would be 4.7% ($42K/900K). The retiree would just trust in history, that future returns will more than compensate for short term setbacks and keep living the same lifestyle. The key question is is there a level of CWR that spells trouble?

Lets look at three historical retiree portfolios and their experience with CWRs. The table below shows 3 retiree portfolios; an Oct 1964 retiree, a Jan 1971 retiree, and a Jan 1975 retiree. It shows their respective CWRs in year 1 (in year 1 the CWR is equal to the SWR), 5, 10, 15, and 30.

Source: Conserving Client Portfolios During Retirement – William Bengen

The 3 retirees had quite different experiences in retirement. All 3 began with SWRs of 4.8%. By year 5, the results started to show some big differences. The 1971 retiree saw a huge jump in their CWR to 8%! Almost doubling. This means in real terms their portfolio almost dropped by half. Talk about gut wrenching. However, those were the darkest days for the 1975 retiree. From year 5 on out to year 30 their CWR dropped back to 4.6% and they lived happily ever after. The 1975 retiree had it really easy. They never saw an increase in CWR through their entire retirement. Then there is the 1964 retiree. The first 5 years were ok then things started to head south. By year 10, the CWR was 7% and by year 15 it had climbed to 9% and by year 30 it reached 13%. Truly frightening. And yet this shows the power of the SWR. Even with the awful performance of the 1964 retiree’s portfolio, the portfolio survived for 30 years and is still going. The experience probably did not meet their expectations and surely they do not have as much wealth to pass on as the other retirees but the 1964 retiree was able to maintain their lifestyle which after all is the main goal of the SWR. But they are awfully close to the complete exhaustion of the portfolio. Its seems to be cutting it awfully close.  Somewhere before a CWR of 13% some changed would probably have been in order. The CWR can serve as a great indicator as to how the retirement plan is going and can be used as a basis to make retirement changes.

In summary, the CWR is a great indicator as to the health of a retirement portfolio. A portfolio can sustain quite an increase in CWRs from the initial SWR. Increases of 25% to 50% is not anything to be overly concerned about. While these results are still clearly dependent on history it seems to me that increases above 50% in the CWR could be used as early warning indicators for a retirement portfolio. A retiree could then make adjustments in lifestyle or adopt other withdrawal methods to assure the portfolio lasts trough retirement.


10 Comments

Mark · February 28, 2011 at 2:30 pm

The Jan75 scenario is a real ballshrinker huh? I can just hear some guy telling his wife “honey, just trust in history!” Yeah, that’s gonna go well =) But, I’m assuming Bengen’s study is just using S&P market returns for each of these 30-year scenarios, is that right? My question is using your growing dividend payer strategy, you would theoretically not care so much about the S&P (or the total market value of your portfolio) as long as everybody is paying you your dividends, no? I would love to see these same scenarios run that way. At least in theory even if your CWR is climbing you could say, “I don’t know about history, but I do think I’m buying stock cheap for later” Wouldn’t you also use the dividend payments themselves (not just the CWR) as a warning that withdrawal adjustments might be necessary? One of the things I like most about your div strategy is that it might help you not to panic in downturns. Focusing on a climbing CWR seems like your back in the mode of fretting over market prices at a particular moment. Did I miss the magic? =)

    libertatemamo · March 1, 2011 at 6:12 am

    Sure is a killer scenario Mark. And you’re right on the mark about a dividend portfolio and living off the dividends. You understand the magic.

    The biggest problem with all retirement studies is that they use only the indexes and then only a max of 3 asset classes (US large cap stocks, US small cap stocks, and US bonds), usually a 60/40 or 70/30 stock/bond mix. Then they assume you always re-invest the income or dividends and just sell off a piece of the portfolio every year to pay for living expenses. This only one retirement model but they only one constantly discusses and uses in all retirement calculators.

    I think there is a better model, the dividend income model.

    Paul

      Mark · March 1, 2011 at 11:50 am

      oops I was referring to Oct64 (the one that spiked to 13% CWR) not Jan75. So what metric do you think should serve as a early warning if you are trying not to pay too much attention to total portfolia value (or the CWR which is based upon that value) ? The only thing I can see is the dividend stream. Are they paying? If not, you’re in trouble and may have to make adjustments. If so, then just hang on through the downturn. Is that what you are thinking?

        libertatemamo · March 2, 2011 at 10:52 am

        I knew which one you mean Mark. And yep, if you’re living off dividends, the portfolio value matters less, but then the dividend stream is all important. In addition, I think an investor has to protect oneself from dividend cuts. The do happen and more often than most dividend investors think. Just using the S&P500, dividends have dropped anywhere from 20% to 40% though history (peak to trough). Outside of the Great Depression, the worst drop has been 20%. So, I think building in a buffer of 20% into the dividend stream plus staying flexible on the spending side should stuff really hit the fan is a better approach. I’ll be posting more on this soon.

J Carroll · March 2, 2011 at 10:21 am

I think the studies are useful, but only to the extent that one truly trusts history will repeat itself. If one happens to be in that 1% that falls in the extreme poor end of the curve, one is 100% broke 30 years later. Getting close to the point of beginning portfolio draw-down, I think I am just going to cash out 4% of the portfolio value on Jan 1 (or whatever arbitrary date chosen), or maybe 1% each quarter, and tell myself “OK, that’s it for the year/quarter.”

    libertatemamo · March 2, 2011 at 11:01 am

    Absolutely J, but what else does one have besides history? If an investor chooses a SWR with a 100% success rate, then the future would have to be worse than the Great Depression, and the 70s stagflation, to run out of money. Could happen, but pretty dire outlook. If an investor chooses anything besides a 100% safe SWR, even a 99% success rate, then yes, they could be in that really unlucky 1% and be broke 30 years later as you mention.

    In your last comment, if you take 4% of your portfolio every year, then you will never run out of money, by definition. That’s called the fixed percentage method. The downside of that method is that your withdrawals vary with portfolio value. If you can make up the gaps in a given year with income from other sources, then you’re golden.

    I’m reminded of a Yogi Berra quote, “Its tough to make predictions, especially about the future”

Will the year 2000 retiree destroy the 4% rule? « Investing For A Living · May 30, 2011 at 11:22 am

[…] way to monitor the progress of retirement portfolios that can signal signs of potential dangers. In this post, I discussed using the current withdrawal rate (CWR) as an indicator of potential retirement […]

The greatest test of the 4% rule in history « Investing For A Living · July 24, 2012 at 7:54 pm

[…] of 4%, the value of the portfolio at the end of 2011, and the current withdrawal rate (CWR). The CWR is an early warning indicator of trouble for retirement portfolios. Now, on to the […]

Revisiting the worst times to retire in history (2012 update) « Investing For A Living · January 29, 2013 at 9:05 am

[…] percentage those withdrawals are of the current portfolio value, or the CWR as I’ve defined before. The 1966 retiree portfolio is highlighted since that was the worst case before. 13 years into […]

Revisiting the worst times to retire in history (2013 update) | Investing For A Living · May 12, 2014 at 10:37 am

[…] is down to $787K and the $61K yearly withdrawal is now 7.7% of the portfolio. CWR is defined here. But that is way better than the 1929, 1965, or 1966 retiree, all of who’s portfolio […]

Comments are closed.