Inflation Linked Bonds: Passive or near-passive

[dealtn]

So my question is at what level of real yield are you comfortable buying. Would negative real yields, and a reduction in purchasing power if held to maturity, still be attractive enough over other assets?

Without those you might want some linkers, as unattractive as their yield might be.

I may have got this wrong, so apologies in advance, but you seem to be looking at yield as an income measure. Of course it is, but my point here is that yield is related to price. High yield equates with low price. Low (or unattractive) yield equates with high price. But linkers have a defined (in real terms) price at maturity, which is 100. So what I am really saying when I say low, or unattractive, indeed negative real yield, is a guaranteed loss of (real) capital, assuming held to maturity. The income is low too, but that is incidental to the argument, not the main thrust of it.

[Alaric]

So what I am really saying when I say low, or unattractive, indeed negative real yield, is a guaranteed loss of (real) capital, assuming held to maturity. The income is low too, but that is incidental to the argument, not the main thrust of it.

There isn't actually a zero coupon linker, but if there were, assuming it was able to be issued at par, the money price from time to time might be 100* (Index at purchase) / (index at issue). If it exceeds that, then the money return for a new purchaser will be below the growth of the index, below inflation in other words. It's something likely to be seen if a government is able to issue money Gilts at a coupon of between 1% and 2% and gives instructions that the economy should be managed with a 2% or higher inflation rate in mind.

[JohnW]

Of course it is, but my point here is that yield is related to price. High yield equates with low price. Low (or unattractive) yield equates with high price.

It’s good to have an educational opportunity with someone who’s worked with the beast, but let’s keep in mind that low yield equates with high price only when interest rates are falling, carrying the bond price up. If a linker is issued with a low coupon of minus 1%, at a low (normal) price of £100, the price will stay at £100 if there’s no inflation and no interest rate drop. If only inflation occurs, taking the price up, then you’ll lose nothing (except for the poor coupon rate) by buying it at this higher price if held to maturity unless deflation takes its value down below your purchasing price (and even then your purchasing power won’t be eroded). In which case you’d have been better off with a nominal bond if the deflation exceeds the inflation. I think.

So what I am really saying when I say low, or unattractive, indeed negative real yield, is a guaranteed loss of (real) capital, assuming held to maturity.

Indeed, but it’s still an inflation protection the like of which nothing else offers, and so we take what we can get, or don’t invest, that’s another option.

[tjh290633]

So if my concern was that inflation might lead to prices doubling in say 10 years, I would be looking at what that would mean for my purchasing power, but also what are the entry and exit prices for my assets under such a scenario? ....A commodity fund might be priced at 100 and increase to 180, and selling would give you 90 pounds of equivalent to today spending power.

MIGHT is important there. How good is the link between inflation and a commodity fund price? Probably not as good as the change in the adjusted face value of a linker (and its coupons) are matched to inflation.

It obviously depends on the fund price movements, which have no correlation with inflation. Likewise in the past they have been prolific providers of distributions, rolled up in the accumulation units. More recently distributions have often been as rare as hen's teeth.

I first bought Ebor Commodity Fund in 1970 at 7/=, or 35p in today's money. I switched into OEIC Accumulation shares in 2002 at 141.8p. Dividends had effectively dried up by 1995, but they have resumed of late. I changed class of units in July last year, so what had been 659p became 78p. Currently 68.88p.

My records show that the RPI in 1970 was 73.1, using the 1974=100 base, and the latest value of 292.7 for June 2020 equates to 561.4 on the same base, allowing for the old base being 191.8 when rebased in 1987.

So we are comparing about 650/35 with 561/73. That's a bit over twice the increase.

TJH

[1nvest]

you seem to be looking at yield as an income measure. Of course it is, but my point here is that yield is related to price. High yield equates with low price. Low (or unattractive) yield equates with high price.

When equities yields are low, prices are high - many still buy equities.

The convexity of bonds has prices move relatively more at/from lower yields, price appreciation potential can be massive, especially with longer dated versions. This Austrian 100 year bond for instance has seen its price around double during the last couple of years https://www.boerse-berlin.com/index.php ... 0000A1XML2 Note also how when stocks dived during the March Covid-19 'dip' how that bonds price spiked heavily upward.

Longer dated bonds can be as volatile as stocks, but over shorter periods can have a inverse correlation, such that combined that can yield a similar reward, with less volatility.

If you opine that you shouldn't hold bonds due to high prices/low yields, as have many since 2008, then you'd also need to be considering the same for stocks at times.

Some do only hold short dated bonds for more stable capital value, and add risk exposure on the stock side. Others might buy something like a 100 year bond, but not with the intent of holding to maturity, but rather for the potential to zag when stocks zig and vice versa.

[1nvest]

My records show that the RPI in 1970 was 73.1, using the 1974=100 base, and the latest value of 292.7 for June 2020 equates to 561.4 on the same base, allowing for the old base being 191.8 when rebased in 1987.

So we are comparing about 650/35 with 561/73. That's a bit over twice the increase.

TJH

IIRC ons data is limited, whereas FRED data goes further back (to 1960) ...
https://fred.stlouisfed.org/series/CPRPTT01GBM661N

[Lootman]

The convexity of bonds has prices move relatively more at/from lower yields, price appreciation potential can be massive, especially with longer dated versions. This Austrian 100 year bond for instance has seen its price around double during the last couple of years https://www.boerse-berlin.com/index.php ... 0000A1XML2 Note also how when stocks dived during the March Covid-19 'dip' how that bonds price spiked heavily upward.

It's true that very low yields magnify the way bond prices can move. In both directions! But the way I was taught to think about these things is that when yields drop and prices go up, that return is really borrowed from the future. A bond held to maturity has a pre-defined payout. If you get a lot of that payout early then by definition you will get less of it later.

Of course you can always sell the bond after a run-up, especially if that bond is now priced over par. But then what do you invest the proceeds in? Most likely all other bonds will now be moreexpensive and sport lower yields. You have reinvestment risk. That is why I think bonds at these very low rates are more a trading vehicle than a long-term investment.

Others might buy something like a 100 year bond, but not with the intent of holding to maturity, but rather for the potential to zag when stocks zig and vice versa.

I have found that US treasuries are the best safe haven when shares are diving. Not only do their yields decline, but the USD usually holds up better than most in a crisis. Gilts can do well but sterling seems to suffer when shares go down. Since the beginning of the sub-prime crisis the pound has declined 35% against the US dollar.

But if you really want leverage then you can do better than a 100 year bond. Try strips or zeroes for maximum volatility.

[1nvest]

Thanks Lootman.

One aspect of Gilts is that price appreciation gains are considered as CGT exempt capital gains (but losses cannot be used to offset against other capital gains). Whilst Gilt interest/income is taxable.

Currency gains are CGT (not income) taxable.

Aren't US Treasuries also open to US withholding taxes?

[Lootman]

Aren't US Treasuries also open to US withholding taxes?

Not for Treasuries, with a W-8BEN:

"Interest (including OID) received by
U.S. nonresidents on obligations of
the U.S. government (e.g., Treasury
bills, notes and bonds) which were
issued after July 18, 1984 (“portfolio
interest” obligations) is generally
exempt from U.S. withholding tax,
as long as a Form W-8BEN has been
provided to the payer."

[1nvest]

Thanks Lootman.

Out of interest, I've just run 30 year constant (yearly) rolled Gilt through a bond price calculator and 1980 to 2019 inclusive that provided a 10.5% annualised reward. Of which capital appreciation made up 38% of the total gains, whilst interest made up the other 62%. Compared to RPI annualised 4.2%

Anyone investing in stocks or bonds since 1980 has had a good deal of 'luck' behind them where it wouldn't have mattered what degree of stock/bond asset allocation you might have held, you'd have done relatively well. Primarily driven by relatively high to very low interest rate transition.

Projecting that bonds wont do well in forward time, could also be considered as a indicator that neither may stocks and anyone basing retirement withdrawal projections, now matter how they might be 'drawn', could be in for a rude awakening.

The 1960 to mid 1970's era comes to mind, where even costless reinvestment of dividends total returns barely broke even in real (after inflation) terms. Such that drawing any, no matter how drawn (dividends/interest/selling shares), was paramount to eating capital.

A transition perhaps from a era/generation of greed/growth, to one of capital preservation.

[tjh290633]

My records show that the RPI in 1970 was 73.1, using the 1974=100 base, and the latest value of 292.7 for June 2020 equates to 561.4 on the same base, allowing for the old base being 191.8 when rebased in 1987.

So we are comparing about 650/35 with 561/73. That's a bit over twice the increase.

TJH

IIRC ons data is limited, whereas FRED data goes further back (to 1960) ...
https://fred.stlouisfed.org/series/CPRPTT01GBM661N

See www.ons.gov.uk/ons/rel/cpi/consumer-pri ... e-data.pdf

That goes back to 1750.

TJH

[dealtn]

If you opine that you shouldn't hold bonds due to high prices/low yields, as have many since 2008, then you'd also need to be considering the same for stocks at times.

That's not my opinion. My words of caution are intended only for (potential) investors to understand them.

Many have the opinion that bonds are low risk. Many have the opinion that buying linkers gives you complete inflation protection. The examples I use (mainly) are buy and hold to maturity to illustrate the point. Buying/Selling for shorter investment periods provide opportunities to both make and lose money in both cash, and real terms.

In general I would agree there may be times when buying equities is higher risk than at other times, but I think that asset class is better understood, and better researched by (potential) investors.

[1nvest]

If you avoid assets in fear of them being high, often today's ceiling becomes tomorrows floor. What is high can/does go higher. Long dated Gilts are priced to a daily 50/50 chance of rising/falling, but yes at current valuations do look like sooner or later they'll take a hit. If you stay out of the market in cash due to valuations being 'high', waiting for a 'correction' to occur, often you'll lose more in missed opportunity costs. 30 year gilt total return for instance since 2009 when many were saying longer dated gilts had only one direction to go (downward) has more than doubled.

I suspect that with all of the money printing going on globally, rising gold prices, lowering yields etc, that sooner or later inflation will hit. When it does, long dated gilt yields wont rise as much as inflation (inverted yield curve) as such gilts are in effect pricing in the average of many years of predicted inflation rate, not the spot inflation rate. So perhaps inflation of 6%, 20 year gilt yield of 3% (up from 0.5%) and a 33% hit/loss. Similarly however gold in seeing even more negative real yields due to high/rising inflation will tend to spike, and where the increase in gold might offset the losses from Gilts. Thereafter as inflation slows and real yields move positive so gold will lose and gilts will gain, perhaps again where the gains in one offset the losses in the other. As a 33% decline requires a 50% gain to get back to break-even, then 50/50 of such assets moving counter to each other yields the positive mean/average, not the 0% compounded. Up overall - despite large losses at times in both bonds and gold.

That volatility could occur tomorrow, or maybe not for a number of years yet. For those that dropped bonds back in 2009 due to yields being 'too high' they've missed out on significant upside gains.

I guess its comfort. During the 1980/90's for instance gold pretty much declined repeatedly. Rebalancing however had you adding more ounces of gold, so what was lost in price appreciation was more than offset by the number of ounces of gold being held having risen by a factor of 5 to 10 times. I suspect it may be something similar in forward time for Gilts, whilst the held gilts might lose, a more stable overall portfolio value due to diversification will tend to see the number of bond certificates being held expand, until such times yields turn around and relatively high value bond certificates start supporting the entire portfolio, buying up other relatively cheap assets such as gold.

[1nvest]

Aren't US Treasuries also open to US withholding taxes?

Not for Treasuries, with a W-8BEN:

"Interest (including OID) received by
U.S. nonresidents on obligations of
the U.S. government (e.g., Treasury
bills, notes and bonds) which were
issued after July 18, 1984 (“portfolio
interest” obligations) is generally
exempt from U.S. withholding tax,
as long as a Form W-8BEN has been
provided to the payer."

Thanks.

US 20 year constant maturity yields since 1954, calculate the yearly total return, adjust that to £'s and discount UK inflation ...



The accumulation real gain (green line) right hand scale is inverted, so as 1950's yields rose up to 1970's high the green line total real value dived considerable, £1 became 18p at the end of 1980 in real terms for total accumulation return, around a -6.6% annualised loss over 25 odd years. But as yields have subsequently declined, so the accumulation gains are massive, rising from that 18p in real terms up to £2.07 at the end of 2019, a whopping 11.5 times real gain over 40 odd years (6.3% annualised real).

The Pound has more than halved over that mid 1950's to recent data range, down from 2.80 dollars per £ to recent 1.3 $ per £ levels, the £/$ was around 2.20 in 1980 so that £ slide has more benefited (occurred since) the post 1980 rise rather than diluting down the 1950's to 1980 decline.

Stocks weren't particularly good across the 1960/1970's decades, at least not in real terms (better in nominal terms but that's not what really counts), near as flat when all of dividends were being reinvested. Short dated bonds would more have reflected/paced inflation, at least in gross terms (less after taxes). Precious metals, silver 1955 to 1971, gold since 1972 after the US broke the $ from gold and the price of gold free-floated, saw that PM achieve a 3.4% annualised real over mid 1950's to the start of 1980.

A Permanent Portfolio that had 25% in each of PM, 20 year US Treasuries, cash (short dated gilts) and stocks, over the mid 1950's to 1980 period would have had stocks flat, bonds down around 6% annualised, gold up around 3% annualised, cash level ... in real terms. So overall marginally negative annualised real. For a pure UK Permanent Portfolio, UK stocks, UK 20 year Gilts, cash and gold ... that annualised 2.3% real over the mid 1950 to 1980 years.

Harry Browne did suggest that for his Permanent Portfolio one should hold domestic stocks and bonds (treasuries/Gilts) and it looks like that was a appropriate/good call.

Across the 1980's/1990's, when the price of gold repeatedly declined, the PP accumulated multiple times more ounces of gold over those years from rebalancing. It looks like that across the mid 1950's to 1980 it would have been accumulating more long dated gilts. In both cases such expansion of ounce of gold being held, and increased number of long dated gilt certificates being held, subsequently paid off - in a big way. Harry did indicate that one of the PP assets will be doing bad at any one time, whilst another will be doing well, and where generally the good asset(s) gains offset the bad asset(s) losses, and more.

I guess you could try and time things, strive to not hold the anticipated 'bad' asset, however that is notoriously difficult to actually do in practice. You're often surprised (and likely to make the wrong choice, making things worse not better).

As a proxy for a Inflation Bond, the Permanent Portfolio could be a reasonable choice.
http://www.harrybrowne.org/articles/InvestmentRules.htm

Rule #11: Create a bulletproof portfolio for protection.

For the money you need to take care of you for the rest of your life, set up a simple, balanced, diversified portfolio. I call this a "Permanent Portfolio" because once you set it up, you never need to rearrange the investment mix— even if your outlook for the future changes.

The portfolio should assure that your wealth will survive any event — including an event that would be devastating to any individual element within the portfolio. In other words, this portfolio should protect you no matter what the future brings.

It isn't difficult or complicated to have such a portfolio this safe. You can achieve a great deal of diversification with a surprisingly simple portfolio.

Just seems to spit out 2.5% to 5% real (moderately) consistently, come what may. Relatively passive, buy a 25 year gilt, and stick with that for 5 years until is 20 years away from maturity at which time roll that into another 25 year gilt series. For the 25% 'cash' cash deposit accounts are a reasonable choice. A low cost FT all share fund/ETF (but personally I prefer the FT250 such as Vanguards VMID), and some gold (join thesilverforum.com and get yourself established with the crew that vacate that board, and you'll likely get spreads down to pretty low levels once you've been seen to trade (buy/sell) honestly a number of times). Review and maybe rebalance once/year, but only if modest drift is apparent, 20% lower, 30% upper are common choices of weightings before triggering actual rebalance trade(s).

More here viewtopic.php?f=56&t=23992 and here viewtopic.php?f=8&t=12360

[Hariseldon58]

@TJH

Regarding the inflation data, this is interesting

https://www.ons.gov.uk/economy/inflationandpriceindices/timeseries/cdko/mm23

I clicked on table rather than chart.

It has the same 73.1 for 1970 and goes up to 2019 with a figure of 1139.3

[tjh290633]

@TJH

Regarding the inflation data, this is interesting

https://www.ons.gov.uk/economy/inflationandpriceindices/timeseries/cdko/mm23

I clicked on table rather than chart.

It has the same 73.1 for 1970 and goes up to 2019 with a figure of 1139.3

That does not seem right, because there has only been one rebasing since 1974, on 1st January 1987. The 73.1 is correct on the 1974 basis, which was at 191.8 when rebased to 100 in 1987.

TJH

[Hariseldon58]

I can’t argue if the figures are correct or otherwise, the ONS is generally a pretty reliable source and the increase in prices by a factor of 10 since 1974 seems to be in the right ball park. I only looked it up because the figures quoted previously seemed rather low.

[Newroad]

Hi Hariseldon58 et al.

I may be missing something, but 73.1 in 1974 through 1139.3 in 2019 is around 6.3% compound (by my admittedly back of a postage stamp calculation).

I am unsure, therefore, what "a factor of 10" means in context?

Regards, Newroad