The 4% Shocker

Research on Safe Withdrawal Rates

Moderator: hocus2004

Post Reply
JWR1945
***** Legend
Posts: 1697
Joined: Tue Nov 26, 2002 3:59 am
Location: Crestview, Florida

The 4% Shocker

Post by JWR1945 »

The 4% Shocker
or
Withdrawing 4% of the Current Balance
or
How 4% turns out to be less than 2.5%!

I duplicated a previous study except that I withdrew 4% of the portfolio's current balance instead of 5%. I had anticipated steadier income streams during times of portfolio stress. Instead, I found that the income stream fell just as far, actually farther, as with 5%.

Portfolios were safer at a 4% withdrawal rate. Their minimum balances stayed higher.

The previous study was Withdrawing 5% of the Current Balance dated Fri Dec 03, 2004.
http://nofeeboards.com/boards/viewtopic.php?t=3152

Here are the conditions.

I have tabulated the amounts withdrawn when one removes 4% of his portfolio's current balance after adjusting for inflation (i.e., using real dollars). These are based upon an initial balance of $100000. Expenses were set at zero.

These data were taken using the latest version of my Deluxe Calculator 1.1V08, which is a modified version of the Retire Early Safe Withdrawal Calculator, Version 1.61 dated November 7, 2002. The calculator was set up with an initial withdrawal amount equal to zero and expenses (which is the sum of the withdrawal percentage of the current balance and the actual portfolio maintenance expenses) set at 4.0%. I used TIPS with a 2% real interest rate for allocations other than stocks.

[Note: dummy values are used for stock returns after 2002. They are set for heavy losses. Thirty-year historical sequences starting in 1973-1980 include dummy data values at the end of their sequences.]

Here are the new tables.

With 50% stocks and 50% TIPS at 2% interest, these are the 5-year rolling averages of the amounts withdrawn at years 5, 10, 20 and 30. Each year's withdrawal is 4% of the current balance.

Code: Select all

Year, 100E10/P year 5 year 10  year 20 year 30

1921   19.61    4260    6003    6152    5276
1922   15.87    4638    6341    6044    5570
1923   12.20    4609    5840    5361    5453
1924   12.35    4994    5534    5147    5623
1925   10.31    5160    5003    4768    5604
1926    8.85    4847    4324    4252    5328
1927    7.58    4823    4374    4219    5526
1928    5.32    4138    4005    3703    5005
1929    3.69    3350    3490    3129    4434
1930    4.48    3189    3572    3118    4625
1931    5.99    3340    3838    3291    5052
1932   10.75    4243    4460    4110    6304
1933   11.49    5148    4882    4967    7473
1934    7.69    4422    3948    4314    6343
1935    8.70    4839    4117    4839    6879
1936    5.85    4033    3452    4325    5846
1937    4.63    3394    3115    4080    5226
1938    7.41    3794    3699    5000    6178
1939    6.41    3534    3548    5029    5952
1940    6.10    3559    3652    5417    6027
1941    7.19    4248    4256    6534    6902
1942    9.90    4696    4717    7234    7318
1943    9.80    4434    4626    6960    6754
1944    9.01    4142    4507    6628    6107
1945    8.33    3946    4520    6426    5527
1946    6.41    3951    4942    6680    5376
1947    8.70    4409    5750    7367    5657
1948    9.62    4510    5842    7219    5215
1949    9.80    4494    5855    6929    4734
1950    9.35    4905    6352    7067    4815
1951    8.40    4710    5781    6107    4233
1952    8.00    4622    5435    5499    3779
1953    7.69    4681    5437    5276    3616
1954    8.33    4915    5548    5112    3647
1955    6.25    4370    4797    4126    3153
1956    5.46    4143    4563    3672    2992
1957    5.99    4297    4681    3595    3182
1958    7.25    4642    4939    3568    3488
1959    5.56    4110    4309    2944    3112
1960    5.46    4196    4252    2897    3221
1961    5.41    4278    4103    2844    3286
1962    4.72    4107    3814    2622    3166
1963    5.18    4338    3957    2712    3394
1964    4.63    4079    3585    2558    3249
1965    4.29    3888    3300    2522    3188
1966    4.15    3760    3154    2570    3256
1967    4.90    3966    3280    2904    3641
1968    4.65    3834    3037    2969    3738
1969    4.72    3778    2937    3105    4062
1970    5.85    3918    3146    3498    4795
1971    6.06    3756    3103    3586    5105
1972    5.78    3560    2959    3574    5165
1973    5.35    3428    2967    3712    5202
1974    7.41    3856    3538    4495    5829
1975   11.24    4311    4104    5188    6044
1976    8.93    3772    3720    4713    4759
1977    8.77    3714    3955    4959    4225
1978   10.87    4124    4658    5865    4155
1979   10.75    4332    4990    6529    3773
1980   11.24    4404    5144    7051    3284
1981   10.71    4241    4970    7075    
1982   13.48    4678    5304    7665    
With 80% stocks and 20% TIPS at 2% interest, these are the 5-year rolling averages of the amounts withdrawn at years 5, 10, 20 and 30. Each year's withdrawal is 4% of the current balance.

Code: Select all

Year, 100E10/P year 5 year 10  year 20 year 30

1921   19.61    4823    8637    9317     8073
1922   15.87    5202    8723    8506     8201
1923   12.20    5064    7447    6917     7763
1924   12.35    5841    6911    6602     8272
1925   10.31    6016    5736    5751     8083
1926    8.85    5554    4690    4969     7689
1927    7.58    5400    4762    4878     8074
1928    5.32    4219    4158    3981     6935
1929    3.69    3020    3317    3049     5724
1930    4.48    2892    3556    3149     6368
1931    5.99    3172    4052    3511     7489
1932   10.75    4735    5236    5048    10811
1933   11.49    5951    5610    6296    13095
1934    7.69    4661    4054    5080    10188
1935    8.70    5375    4382    6159    11718
1936    5.85    4041    3352    5187     9154
1937    4.63    3115    2886    4777     7770
1938    7.41    3827    3887    6771    10433
1939    6.41    3412    3605    6768     9805
1940    6.10    3406    3700    7482     9893
1941    7.19    4281    4472    9541    11651
1942    9.90    5149    5359   11476    13106
1943    9.80    4801    5305   11033    11839
1944    9.01    4308    5108   10244    10155
1945    8.33    3967    5132    9763     8690
1946    6.41    3675    5444    9606     7677
1947    8.70    4574    7275   11831     8730
1948    9.62    4962    7823   12054     8021
1949    9.80    5022    7951   11520     6956
1950    9.35    5447    8513   11257     6733
1951    8.40    5203    7493    9151     5625
1952    8.00    5152    6937    7922     4813
1953    7.69    5219    6884    7387     4457
1954    8.33    5661    7170    7108     4549
1955    6.25    4705    5727    5098     3612
1956    5.46    4270    5232    4181     3271
1957    5.99    4495    5430    4007     3580
1958    7.25    5134    5998    3991     4193
1959    5.56    4267    4880    2947     3527
1960    5.46    4390    4769    2853     3706
1961    5.41    4552    4537    2789     3839
1962    4.72    4259    4027    2446     3608
1963    5.18    4637    4260    2571     4017
1964    4.63    4231    3667    2347     3759
1965    4.29    3909    3203    2269     3620
1966    4.15    3677    2948    2306     3700
1967    4.90    4003    3124    2791     4420
1968    4.65    3776    2735    2874     4591
1969    4.72    3654    2546    3048     5186
1970    5.85    3912    2855    3710     6787
1971    6.06    3690    2829    3895     7532
1972    5.78    3365    2620    3863     7603
1973    5.35    3051    2542    3971     7433
1974    7.41    3637    3341    5352     8890
1975   11.24    4518    4385    6996     9833
1976    8.93    3707    3783    6069     6792
1977    8.77    3586    4115    6517     5538
1978   10.87    4184    5276    8429     5297
1979   10.75    4398    5730    9751     4395
1980   11.24    4533    6068   11102     3532
1981   10.71    4329    5841   11294    
1982   13.48    5189    6668   13124    
Remarks

Withdrawing a constant percentage of the current balance prevents a portfolio's balance from falling to zero. The buying power of the amounts withdrawn drops quite a bit.

All initial balances were $100000 in this study.

The mid-1960s, especially 1964-1966, were the worst case.

Previously, with 50% stocks, the buying power dropped between $2600 and $2700 at year 20 before rising later. With 80% stocks, there was a lot of variation. Previously, the 5-year average withdrawal amount at year 20 fell to $2375 during the worst case (1965) sequence.

This time, when withdrawing 4% of the portfolio's current balance, the buying power of a 50% stock portfolio fell between $2500 and $2600 at year 20. The worst case sequence with 80% stocks was still in 1965. The 5-year average withdrawal amount at year 20 was $2269.

There are many examples of higher returns at year 20, but not when starting with stocks at high valuations.

Withdrawing a percentage of a portfolio's current balance instead of its initial balance extends its lifetime indefinitely, but only at a cost. The buying power drops substantially by year 20. We are talking about drops of 40%.

The old adage, which I am guilty of repeating, that withdrawing 5% of a portfolio's current balance is roughly equivalent to withdrawing 4% of its initial balance is false. Even withdrawing 4% of a portfolio's current balance reduces one's buying power in later years. It gives the illusion of being a safer, simpler method. It doesn't work out that way. This approach may reduce one's buying power so much that he abandons his original plan. He may withdraw more out of necessity.

Have fun.

John R.

NOTE: Tables with balances follow.
JWR1945
***** Legend
Posts: 1697
Joined: Tue Nov 26, 2002 3:59 am
Location: Crestview, Florida

Post by JWR1945 »

With 50% stocks and 50% TIPS at 2% interest, these are the 5-year real balances at years 5, 10, 20 and 30. Each year's withdrawal is 4% of the current balance.

Code: Select all

Year 100E10/P at year 5 at year 10  at year 20 at year 30

1921   19.61    136124    162102    155160    164314
1922   15.87    130722    122980    127878    158707
1923   12.20    135481    109106    117874    148209
1924   12.35    165191    134313    127630    147940
1925   10.31    140996    113568    122507    161098
1926    8.85    119085    128739    130261    166262
1927    7.58     94077    137996    107161    158523
1928    5.32     80532     99856     87773    128465
1929    3.69     81307     87540     74163    123068
1930    4.48     80547     93147     84645    132369
1931    5.99    108107     95718    101364    145575
1932   10.75    146684    103983    129051    187802
1933   11.49    123995    108036    135839    182073
1934    7.69    107666     95025    110146    159758
1935    8.70    115642    107871    141851    180122
1936    5.85     88540    101183    129147    151028
1937    4.63     70889     77655    114875    125935
1938    7.41     87129     87900    128650    155627
1939    6.41     88259     84718    140584    148088
1940    6.10     93280     90872    142108    135502
1941    7.19    114280    105899    152088    142986
1942    9.90    109544    124107    180608    163677
1943    9.80    100884    125735    168530    165800
1944    9.01     95988    115913    168123    134563
1945    8.33     97419    131500    166979    107427
1946    6.41     92667    127637    149263    104875
1947    8.70    113295    147930    162173    122716
1948    9.62    124633    146360    177050    118111
1949    9.80    120757    165943    174801    116667
1950    9.35    134984    156382    149112    107680
1951    8.40    137738    143616    135020    102255
1952    8.00    130571    145525    131883     88910
1953    7.69    117433    134036    131864     93054
1954    8.33    137419    145043    116090     98970
1955    6.25    115852    126980     81693     82405
1956    5.46    104267    116943     82166     79940
1957    5.99    111453    109628     82955     89533
1958    7.25    114138    120969     80699     90869
1959    5.56    105548    105338     70306     81583
1960    5.46    109605     95351     68857     85456
1961    5.41    112157     94015     71200     81212
1962    4.72     98362     90626     61096     84781
1963    5.18    105984     98380     69424     88146
1964    4.63     99801     80038     68235     84167
1965    4.29     86995     64336     64896     77936
1966    4.15     83824     70262     68358     85892
1967    4.90     92135     75670     81670    100218
1968    4.65     92825     66711     75118    106703
1969    4.72     80198     66743     77449    118563
1970    5.85     73953     72214     89623    134673
1971    6.06     83820     75733     86382    128401
1972    5.78     82130     67415     93551    113465
1973    5.35     71867     70568     89598     95459
1974    7.41     83223     85253    105158     96294
1975   11.24     97649    100872    121140    100177
1976    8.93     90352     97290    122246     78618
1977    8.77     82084    107929    132441     68547
1978   10.87     98192    112603    159948     65904
1979   10.75    102440    116041    177642     57409
1980   11.24    103300    124107    186491     50901
1981   10.71    107679    114061    169545    
1982   13.48    131487    138767    168307 


With 80% stocks and 20% TIPS at 2% interest, these are the 5-year real balances at years 5, 10, 20 and 30. Each year's withdrawal is 4% of the current balance.

Code: Select all

Year 100E10/P at year 5 at year 10  at year 20 at year 30

1921   19.61    170941    232518    225974    269590
1922   15.87    160654    147410    165714    255638
1923   12.20    169409    122515    146306    230325
1924   12.35    228816    167777    165671    229656
1925   10.31    178273    128698    155643    260460
1926    8.85    136022    154916    170667    273195
1927    7.58     91756    172179    123763    253487
1928    5.32     72319    102163     90819    182770
1929    3.69     73324     83974     70490    171685
1930    4.48     72191     92859     86950    193124
1931    5.99    113890     97186    115944    225650
1932   10.75    187648    112417    173419    344440
1933   11.49    141266    119418    187996    327651
1934    7.69    114524     98745    136881    269927
1935    8.70    128629    120936    202381    327123
1936    5.85     85333    110167    176351    251239
1937    4.63     59909     71880    147223    188849
1938    7.41     84534     88896    178901    268742
1939    6.41     86222     83942    204450    248700
1940    6.10     94019     93636    207975    215095
1941    7.19    129103    119301    232184    234786
1942    9.90    119983    154264    306395    292697
1943    9.80    105160    157427    274373    298630
1944    9.01     97357    138621    273359    212601
1945    8.33     99592    167345    270492    146540
1946    6.41     92408    160075    228053    141392
1947    8.70    128571    204817    262728    183678
1948    9.62    149702    201247    302309    172360
1949    9.80    142385    243561    296275    169047
1950    9.35    168030    222110    229713    149141
1951    8.40    173227    194620    196801    137749
1952    8.00    159302    198617    189738    109977
1953    7.69    134432    174286    189695    117930
1954    8.33    171057    197198    153368    129934
1955    6.25    132185    161637     87568     98443
1956    5.46    112349    142466     88328     94023
1957    5.99    124679    128274     89679    111992
1958    7.25    129647    150218     85646    114756
1959    5.56    115282    121643     69407     97491
1960    5.46    122281    103423     67147    104803
1961    5.41    126806    101121     70778     96532
1962    4.72    102883     95529     55371    103039
1963    5.18    115867    108841     67665    109700
1964    4.63    105518     77773     65890    102117
1965    4.29     84578     54176     60904     90426
1966    4.15     79745     62000     65997    104900
1967    4.90     92852     69912     87307    133908
1968    4.65     93936     57014     76393    147409
1969    4.72     73706     57057     80145    173348
1970    5.85     64054     64925    101334    212965
1971    6.06     77748     69994     95462    197072
1972    5.78     75294     57963    107861    161194
1973    5.35     60695     62168    100789    121500
1974    7.41     77412     84720    131301    123352
1975   11.24    101360    112420    166913    132230
1976    8.93     90027    106448    169195     89649
1977    8.77     76982    124881    191537     71465
1978   10.87    102428    133989    258548     66823
1979   10.75    109441    140463    303813     53192
1980   11.24    110912    156078    328018     43548
1981   10.71    118240    136386    281556    
1982   13.48    162221    186087    278099    
[Note: dummy values are used for stock returns after 2002. They are set for heavy losses. Thirty-year historical sequences starting in 1973-1980 include dummy data values at the end of their sequences.]

Have fun.

John R.
hocus2004
Moderator
Posts: 752
Joined: Thu Jun 10, 2004 7:33 am

Post by hocus2004 »

"It gives the illusion of being a safer, simpler method. It doesn't work out that way.."

And the hits just keep on coming! This is why I say that Numbers Matter.

I've never been super keen on strategies that call for a lower take-out after bad return years. That's what the 5 Percent Strategy is, isn't it? By taking 5 percent of the current-day portfolio rather than 4 percent of the starting-date value, you are reducing your take-out in years in which you suffered a portfolio loss. I certainly had no idea about the numbers you put forward here. It just seemed to me that taking a lower withdrawal after already taking a hit to your portfolio is like closing the barn door tight after the horses have run away. The time to do your safety assessments is before you lose accumulated capital. When you lose accumulated capital, your options are narrowed.

A strategy that seems to me to make sense is the one where you plan on taking low withdrawals in the early years of retirement and then to increase them in the event that you don't suffer any big hits in the early going. It is the early years in which portfolio losses do the greatest damage. It seems to me that once you get past the danger zone, it would be OK to loosen up the reins on spending.

The downside of this strategy is that it calls for less spending when you are younger and more when you are older. Lots of people want to do it the other way around. Still, I think it can be an effective strategy in many circumstances. For example, someone might construct a plan where he took out less from his portfolio for the first eight years but in which he planned to work part-time for eight years and thereby generate a separate income stream that would permit his overall spending to go to the level he wanted on a long-term basis. By the time he left the part-time job, he would know whether the higher take-out from the portfolio was safe or not, and he could then proceed to up his take-out without taking on an unacceptable level of risk.
JWR1945
***** Legend
Posts: 1697
Joined: Tue Nov 26, 2002 3:59 am
Location: Crestview, Florida

Post by JWR1945 »

hocus2004 wrote:A strategy that seems to me to make sense is the one where you plan on taking low withdrawals in the early years of retirement and then to increase them in the event that you don't suffer any big hits in the early going. It is the early years in which portfolio losses do the greatest damage. It seems to me that once you get past the danger zone, it would be OK to loosen up the reins on spending.
In terms of the conventional approach (which does not include confidence limits and which fails to include valuations), reducing one's withdrawal rate by 0.3% (roughly from 4.0% to 3.7%) would have extended one's historical retirement survival period from 30 years to 50 years. This is compared to the data which I have just presented. Withdrawing 4% of one's current balance allows the withdrawals to fall below 2.6% under stressful conditions.

Repeat: reducing steady withdrawals from 4.0% of the initial balance to 3.7% of the initial balance would have extended a 30 year retirement to 50 years. Withdrawing 4.0% of the portfolio's current balance is a poor alternative. Withdrawals can end up falling below 2.6% of the portfolio's initial balance.

[All of these amounts include adjustments to match inflation. That is, withdrawals are in real dollars.]

Hocus2004 is right about getting past the danger zone. This shows up in a very natural, easily understandable manner. Safety occurs when one's portfolio has grown substantially. Withdrawals are a smaller, typically a much smaller, percentage of the current balance than they are for the initial balance.

Have fun.

John R.
hocus2004
Moderator
Posts: 752
Joined: Thu Jun 10, 2004 7:33 am

Post by hocus2004 »

"Hocus2004 is right about getting past the danger zone. "

We know each other pretty well, JWR1945. Please feel free to refer to me as just plain old "hocus."
Post Reply