Exploring drawdown options - take from cash vs equities first?

Audaxer

jamesd said:

Linton said:

The problem is that none of the backtesting data includes close to zero bond rates. For much of the past 100 years bonds provided useful returns reducing the impact of crashes. Removing emotion is certainly a good thing, but calmly driving around with your eyes closed is not to be recommended.

There were some such periods in the US data. That's why I've been able to write about cash being shown to be a better choice during low interest rates, low inflation times. People backtested it with the historic data for those periods.

Linton said:

jamesd said:

...
a. first use income units so you get cash distributions instead of the reinvestment from accumulation units.

I agree with this general strategy and use it myself (with tweaks).  Interestingly it does support the use of some income funds despite the preference on this forum for a "total return" strategy.

In practice I wouldnt necessarily sell all the excess equity as it seems sensible to constrain the total amount in cash because of  inflation risk.  So equity is used for money for which there is no specific use.

Once you get to (c) part 3 surely you are well on your way to disaster.

In not selling the equity you'd be making a mistake, I think. You're in a bull run or at least bull market situation and what comes next, eventually, is a crash or correction. But you are allowed to buy fixed interest with the excess, it doesn't have to be all or even at all kept in cash.

I know it makes sense to look at Total Return for growth funds, but with Income funds/ITs I'd be reluctant to sell capital as it would mean lower future dividends. Say I had £10k invested in an equity income IT that currently pays a dividend of £400 per annum rising with inflation. If that IT's capital value grows by say 20% in the next year and is now worth £12k, it would be good to top slice the £2k capital growth as income, but I would now have less units and therefore lower future dividends from that IT. So would I be making a mistake by not selling some of the capital in that example?

sheslookinhot

Make sure you use up your personal tax allowance.

JohnWinder

Right now, I would also question the efficacy of an automatic 20% allocation to bonds (except index linked).

Also I would see the 20% in bonds as a potential problem.  Bonds are there to provide some diversification from equity.  However safe bonds are at a price where the effective interest rate has been squeezed to near zero which means there is very little room for increased capital value to mitigate an equity crash.  Indeed there is the possibility of both equity and safe bonds falling in price at the same time.

Thanks all, I recognise that currently bonds aren't likely to perform so I may consider just holding that 20% as cash as well,

At the risk of reading too much into these statements, they all seem to be pointing to the general view that bonds are on the nose just now. As for anything else I think it's useful to try to quantify this problem. Wisely, elsewhere in this thread we see equity crashes being quantified at 50%, but to avoid irrational decisions I think we need to quantify how badly we think bonds will let us down in the short/medium/long term future - whichever is important to you. Because if they're going to crash and burn by 20% we should respond very differently than if it's going to be 1%.

So, would anyone like to quantify the hazards investment grade bonds face over the next 5-7 years? Or bluntly, how much are they going to fall in the scenarios you're envisaging when we warn each other away from bonds?

Spoiler: here's what happened to some bond funds when the central bank rate rose 12% in three years, almost nothing. https://www.bogleheads.org/forum/viewtopic.php?p=6287982#p6287982

Linton

JohnWinder said:

Right now, I would also question the efficacy of an automatic 20% allocation to bonds (except index linked).

Also I would see the 20% in bonds as a potential problem.  Bonds are there to provide some diversification from equity.  However safe bonds are at a price where the effective interest rate has been squeezed to near zero which means there is very little room for increased capital value to mitigate an equity crash.  Indeed there is the possibility of both equity and safe bonds falling in price at the same time.

Thanks all, I recognise that currently bonds aren't likely to perform so I may consider just holding that 20% as cash as well,

At the risk of reading too much into these statements, they all seem to be pointing to the general view that bonds are on the nose just now. As for anything else I think it's useful to try to quantify this problem. Wisely, elsewhere in this thread we see equity crashes being quantified at 50%, but to avoid irrational decisions I think we need to quantify how badly we think bonds will let us down in the short/medium/long term future - whichever is important to you. Because if they're going to crash and burn by 20% we should respond very differently than if it's going to be 1%.

So, would anyone like to quantify the hazards investment grade bonds face over the next 5-7 years? Or bluntly, how much are they going to fall in the scenarios you're envisaging when we warn each other away from bonds?

Spoiler: here's what happened to some bond funds when the central bank rate rose 12% in three years, almost nothing. https://www.bogleheads.org/forum/viewtopic.php?p=6287982#p6287982

The 12% rise in 3 years example is not that applicable to the current situation.  The behaviour of safe bonds is very dependent on the time to maturity.  An obviously short term emergency rise in interest rates wont affect long term rates very much.  Long term interest have fallen steadily since the Great Crash and are now at very low values.

To give some rough numbers as to the current level of risk....

Have a look at https://www.fixedincomeinvestor.co.uk/x/bondtable.html?groupid=3 which gives a list of UK gilts.  The VGOV UK Gilt index fund has an average time to maturity of just over 13.4 years.  So choosing 2 gilts with maturity dates in a similar tome frame:

A ) A recent bond issued at 0.625% maturing in 13.75 years current price 91.2p Yield to maturity 1.326%
B ) An old bond issued at 4.5% maturing in 12.9 years current price 138.7p Yield to maturity 1.236%

Using the https://www.fixedincomeinvestor.co.uk/x/yieldcalc.html calculator:

If the market rates rose by 1% today the prices would be:
A) Price 80p
B ) Price 125p

So we are talking about 10% drop in capital value for a 1% increase in interest rates.

MarkCarnage

It might be helpful to think of cash as simply a rolling zero duration bond when making this comparison....yes this is a simplification, and ignores credit risk but it does highlight that a key factor here is the duration of the bonds in question. 

I suspect that inflation is likely to be the more insidious enemy here than rate rises per se. Hence the question is perhaps real fall rather than nominal fall. Unthinkingly holding significant cash for long periods carries the same risk too of course. 

Answering the question, nominal falls of < 10% would be my guesstimate, but over 2-3 years that could easily translate into real falls of double that, though money illusion will make that less of a shock. There is very little coupon cushion right now too. 

EDIT: My answer is in the context of short/medium duration investment grade which fits the time horizon envisaged in the notional asset allocation for drawdown. The answer is similar to that given by @Linton above using a yield calculator which probably assumes an instantaneous parallel shift in the yield curve of 1%.

OldScientist

jamesd said:

Regrettably, ERN falsely claimed to use the Guyton-Klinger rules and have refused to correct their major mistakes after requests from many people. The result was a performance mess as they created their own poor rules instead of using the real set.  Wade Pfau produced graphs showing how something closer to real Guyton-Klinger works, though he also didn't use the Portfolio Management Rule, and also explains two of the changes from Guyton to Guyton-Klinger.

One potential drawback from G-K (like any dynamic withdrawal approach) is that the withdrawals can become small - e.g. in the example on the link to Pfau's work you provided, the income falls to about 2% in real-terms (i.e. half the initial withdrawal) so that degree of flexibility has to be built into a retirement plan. For a UK based portfolio, in the worst case, the withdrawal falls to about 1.5% (with the same conditions as for Pfau, i.e. 50/50 and initial withdrawal of 4%).

To remain on topic, the results presented in G-K's 2006 paper indicate that the portfolio management rule had a relatively small effect on the income (peaking at 6-7%) compared to the the other rules. It is interesting that McClung's backtesting (Figure 10 - see http://livingoffyourmoney.com/wp-content/uploads/2016/05/LivingOffYourOwnMoney_eBook_FirstThreeChapters.pdf), suggests that the PMR rule shows a 'modest improvement over traditional rebalancing' (I found his footnote dryly amusing).

JohnWinder

Yes. But to get that 10% drop in capital value you have to imagine 10 year bond yields rising 1% in an instantaneous step jump. Looking at the graphs they don't seem to do that, not that they couldn't; but even if the central bank made a 1% step change in the overnight lending rate (is anyone picturing that happening now?) it wouldn't mean the 10 year yield which of course is the relevant figure would change 1%.

But from the discussion in that link, any increase in bond yield progressively plays into the bonds in the fund paying more which is what all bond holders want, so that within a short few years the fund value has recovered its lost 10%. Part of the mechanism for this is explained in that link: 'As an example, if I buy a 5 year bond today the yield is 1.16% (for simplicity let's say the price was $1000 and the coupon was exactly 1.16%). If I hold it for a year, and interest rates don't change at all, I would now hold a 4 year bond with a .96% yield - that's the market rate for 4 year bonds. Since yield and price move inversely, the price of my bond is now higher. Same coupon as before, but higher price. The combination of the 1.16% coupon, ~$1001 price, and $1000 face redepmtion value, mathematically equals .96% yield.

If interest rates rose 0.1%, the 4 year yield would now be 1.06% - still lower than when I bought it, so the price would still be over $1000. Plus I have the coupon.

Since you are doing a ladder holding to maturity - this doesn't apply. But for most real bond funds roll yield is a significant part of returns. Rolling down the yield curve to shorter durations = price increases (all else equal). This is especially true where the curve is steep - less than 10 years of duration.'

A lot of our angst over lower interest rates, potential rises and bond damage, is to do with how quickly rates rise. Slow rises are hardly noticeable in bond fund prices, for the above reasons. Quick rises have visible price effects (down), but they're short lived compared to the holding period for the bond fund by the investor. One shouldn't hold long duration funds if your needs are short term, but if you're a thirty year investor holding a bond fund with a 8-10 year duration, you really shouldn't be worried about interest rate rises; rather, you should be embracing them.

MarkCarnage

I agree that this is very unlikely to be some instantaneous shock. 

Institutional demand (from DB pension schemes, life insurance companies) is pretty hard wired to mop up supply of gilts and corporate bonds for the foreseeable future due to both the maturity of DB schemes and legislative requirements. They have limited price sensitivity, though could invest globally and hedge the currency risk. 

And yes, the roll rate will diminish the impact for an open ended fund. However, coupons are at historic lows so income flows will not have the same dampening effect that they have in the past. 

Should inflation rise even a bit on a sustained basis (or perhaps more accurately, should recent rises not be reversed), then there will be an insidious negative real return effect, perhaps modest, but possibly persistent. That's probably the bigger risk though less noticeable. Another risk would be to mismatch your bond duration to your liabilities. In other words, don't hold a 20 year duration fund if your bonds are intended to cover the next 3-5 years of spending. At current yields, I think there is a much better argument for doing the opposite mismatch, in other words holding cash not 5 yr duration bonds. 

JohnWinder

I think there'd be plenty of agreement that the prospects for 5yr duration bond funds are not so fantastic that choosing cash instead would be a poor choice, so yes, cash is a suitable alternative to short duration bonds although in theory there should in the long run be some better returns for taking interest rate risk. But since we're not talking about a big yield difference, or a big part of a portfolio, it's not something to agonise over.

But it's not just now that spending needs in the next 3-5 years should be covered by cash rather than a 5 yr duration bond fund. Those spending needs have a duration of about 2.7 years, so a 5 yr duration fund is wrong right from the start, and only gets wronger as the spending period runs down while the fund duration stays the same. As you say, keep the spending duration less than the bond duration to protect against interest rate changes.

Linton

I would be quite happy to withdraw the instantaneous comment.  It was an attempt to get around the problem that we are dealing with a bond fund here rather than following an individual bond down.   It is reasonable to assume that the average bond maturity in VGOV will stay at 13.4 years but the 13.4 year average bond in a year's time will be one which was 14.4 years to maturity the year before and so would have suffered a different % fall.

The aim was to get an approximate value for the degree of risk.  If anyone has a better way of getting at this it would be good to see the results.

It could be better to have a bond fund that has say a 5 year duration, but unfortunately they dont seem to be available in the UK so it would seem that we are stuck between broad funds that include very long duration bonds whose value will suffer much more severely and cash.

For an insight into the effect on long term bonds....

There has recently been issued a 50 year gilt with a coupon interest rate of 1.625%.  It is currently valued at £1.19 with a yield to maturity of 1.127%.  Changing the YTM to 2.127 puts the price at 80, a drop of about 30%.