New discussion on cash buffers, portfolio management etc

Research on Safe Withdrawal Rates

Moderator: hocus2004

Post Reply
peteyperson
**** Heavy Hitter
Posts: 525
Joined: Tue Nov 26, 2002 6:46 am

New discussion on cash buffers, portfolio management etc

Post by peteyperson »

I have been asked to post some points from a recent thread, that later became sunk when SWR semantics dominated the discussion instead. Therefore to some, the discussion below may seem like repetition. My apologies but it was requested of me.


To my way of thinking, any safe withdrawal analysis based on the past only takes you so far. No one has 100% safety during FIRE. Even $100m could get wiped out if the economy collapsed, inflation hit 50%. Your cash reserves would eventually dwindle, if you owned property pretty soon people wouldn't be able to pay your rent etc. So I start from that position and work back from there.

Historical analysis can give you a guide figure, what might work if you started FIRE on any year for x number of years and had that sequence of returns in that order. You can play with the order of the returns, but essentialy that's what it is. It is a fancy smancy way of trying to answer the question of what can I take out each year or in other words, how much do I need to FIRE in relative safety, what's my magic number? What frustrates some is that there is no one magic number, no single answer. It is more of an educated guess, backed up by the past performance, your approach to asset allocation and how much risk you wish to take on.

Given that opinion, I started looking at what I could add to the approach without needing the kind of mathematical skills that others possess. Bogle started me on the road to thinking about simplifying problems to get to simple solutions, overcomplicating things can hide problems in the approach or calculation. You get bogged down in the issues and miss things because it was hidden in a row of complex calculations & known/unknown assumptions.

I see various components to a FIRE portfolio...

I think a stock component is critical to boost the potential return. On occasion there are assets that can outperform stocks but knowing which before the period of outperformance and/or underperformance of stocks is another question altogether. Adding stocks boosts volatility however and so the potential of higher returns is a mixed blessing.

A range of other options exist for a more liquid component. I feel in order to counterbalance the risk of substantial underperformance of stocks over the medium term (1-15 years), investments that can be turned into cash without penalty is a critical component that acts as a solid counterbalance. It does reduce the overall return, but in turn it lowers the risk of FIRE failure by restricting as much as possible the need to sell stocks when they're undervalued. SWR studies cannot take into account timed sales instead of the standard reverse dollar cost averaging approach. They also tend to look at bond return rather than a mix of cash, bonds and other instruments which collectively probably have a lower return. Some are also more liable to investment fees and yearly taxation on income (though I concede there are more ways to avoid this for US investors).

A balanced portfolio of stocks (preferably in a low cost index fund that doesn't carry with it potential issues with star manager changes, high stock turnover rate causing higher taxes etc), a mix of low cost bonds (preferably indexed too) and cash, lowers return but brings a level of fiscal responsibility to the need to preserve capital. Often growth is thought to be important during retirement. I think preservation of capital is a more critical issue and limiting your downside is more important than creating heady room for a wonderous upside. You're not in a position to substain heavy losses and your asset allocation & withdrawal rate should reflect that reality.

I use myself as an example. My FIRE budget includes a category for recreation. This includes going out money, 'Mad Money' for anything I like (books, dvds, etc) and an annual 2 week vacation. It forms 25% of my FIRE budget. Furthermore, I can put off home improvement projects and some other expenditure. All told, I have the flexibility to cut back 10-30% should I need to, for a limited period. The asset allocation reduces the withdrawal rate to something near 2%, but lowers risk considerably and provides some peace of mind to sleep at night. Being able to lower spending even during times when you are spending cash resources and not selling underpriced mutual funds allows added flexibility & further peace of mind during poor periods. It can help cover times when you think you may come out of the down period having had lower overall growth on the portfolio and find you need to have spent less to reduce the impact of that. I think this is far superior to ping pong discussions on whether SWR studies are right and to what extent.

On the subject of valuation, I agree that valuation is taken care of with swr studies because they compare every period whether over or undervalued at the time. However, the more simplified method relies on starting from the same point for all and having clarity. As such, any overvaluation on your stock market component should be revalued down to the historical P/E ratio. This prevents you comparing apples and oranges or deluding yourself. If you are undervalued coming up to FIRE, I would take the value of your stock portfolio as is with no correction. It would be wrong to rachet the value up. You would have to retire on the supressed valuation and if things correct later then you may want to readdress the issue. The point here is again, caution, reduce risk, underestimate rather than overestimate. Not accepting a high valuation as gospel prevents you FIREing on a balloon that goes pop and takes your FIRE dreams with it.

It is perfectly possible for overpriced market to fall to below average valuations. This has happened on several occasions where the market overcorrects. Not adjusting for a high valuation before looking at what your asset allocation can deliver in income for you, is shortsighted. If you start from an undervalued market, the downside is limited to begin with and the statistical likelihood is that your portfolio will correct upwards. You may allocate more to stocks during the Accumulation phase, but reducing the allocation as you get nearer to FIRE will limit the risk of your portfolio becoming undervalued close to quitting work.

My workings are:
(Siegel data)

Assumption: 3% inflation. No capital gains tax included.

Scenario 1 - Historical market return, less P/E increase component reducing historical return from 10.5% to 8.7%

60 / 40 Stock / Cash split

4.5% growth

4.2% dividends, taxed @ 34%, down to 2.77%

Stock return: 4.04% (60% of portfolio)

Cash return: 1.376% (40% of portfolio)
3.44% via INGdirect.co.uk (after taxes)

Inflation: 3%

Net Total: 2.41%
(including taxes on cash interest & dividends)

Net Return: 3.6%
(excluding all taxes)


Costs taken off: UK index funds @ 0.53% (0.2% in the US wouldn't change the % by that much).

I would plan to have some money in bonds but move a sizeable portion of the remainder should interest rates in the UK reach US levels where cash investments deliver 1% pre-inflation, pre-tax. In most circumstances though, cash delivers inflation matching returns with safety of principle vs. bonds variable return dependent on the present interest rates & included investment costs. This makes bonds a higher risk category.


Scenario 2 - Below historical return - 7% gross market return

60 / 40 Stock / Cash split

4.5% growth

2.5% dividends, taxed @ 34%, down to 1.65%

Stock return: 3.372% (60% of portfolio)

Cash return: 1.376% (40% of portfolio)
3.44% after taxes via INGdirect.co.uk

Inflation: 3%

Net Total: 1.74%

If the dividends remain high but price growth is less, then the tax burden on dividends is higher, pushing the w/d rate down to 1.4%. I think however in most circumstances businesses would tend to be cautious, cut back dividends and see how the economy is. They are unlikely to grow less but maintain dividends on average.

If I can afford to, I would like to take a 2% w/d rate. This projects lower than the historical returns on the lower risk allocation selected. The cash buffer adds protection, the flexible budget more still. Index funds lower risk and avoid higher tax liabilities with high turnover rates and chasing the hot fund manager for that extra couple of percentage points who might jump ship anyway costing more in taxes, sales loads in and out etc.. It allows for a very long term buy and hold of the total market index keeping a close watch on all investment costs, taxes and allowing movement. Overall, lower risk of default, lower risk asset allocation, cash to sleep comfortably at night. And no endless discussions on the validity or lack thereof of the latest SWR study as the panacea for all known ills relating to FIRE..

Petey
Last edited by peteyperson on Thu Jul 24, 2003 9:43 am, edited 1 time in total.
peteyperson
**** Heavy Hitter
Posts: 525
Joined: Tue Nov 26, 2002 6:46 am

Post by peteyperson »

Excluding taxes, the historical return on the 60/40 allocation would be 3.6% gross.

2.4% is after paying 20% taxes on cash interest and 34% interest on dividends. Your millage may vary.

Alternating the split up to 70 / 30 instead of 60 / 40 will increase the gross to 4% and the net to 2.8% (UK taxes). The irony of the 4% is not lost on me..

Petey
peteyperson wrote:IMy workings are:
(Siegel data)

Assumption: 3% inflation. No capital gains tax included.

Scenario 1 - Historical market return, less P/E increase component reducing historical return from 10.5% to 8.7%

60 / 40 Stock / Cash split

4.5% growth

4.2% dividends, taxed @ 34%, down to 2.77%

Stock return: 4.04% (60% of portfolio)

Cash return: 1.376% (40% of portfolio)
3.44% via INGdirect.co.uk (after taxes)

Inflation: 3%

Net Total: 2.41%

Costs taken off: UK index funds @ 0.53% (0.2% in the US wouldn't change the % by that much)
Post Reply