How to Protect Against Inflation
How to Protect Against Inflation
Andrew Barrelle has spent the past 10 years thinking about inflation on behalf of some of the world’s largest banks. He has been kind enough to share his thoughts and experience with Bristlemouth. Please note that the opinions provided below are Andrew’s own personal views and not necessarily those of his employer, Royal Bank of Scotland.
Hi Andrew and thanks for sharing your thoughts with us. Firstly, what’s your role at RBS?
I look after the AUD Inflation Trading business at RBS (we were formerly ABN Amro – as of four weeks ago officially part of RBS in Australia). We buy, sell and originate Inflation Linked Bonds and Inflation Linked Swaps and are one of the largest participants in this niche market. There are $20bn of existing Inflation Linked Bonds in the Australian market, with much larger markets in the US, UK, Europe, and even Japan (the land of deflation).
You’ve read Gareth Brown’s views about the potential inflation risks arising from government stimulus. What’s your view?
I spend every working day thinking about inflation, and there has never been a time in the last 15 years where the outlook for actual inflation is more uncertain. On one side you have significant deleveraging and deflationary forces, and on the other an unprecedented government response, including quantitative easing in UK and US (aka the electronic printing press). On balance, I favour higher inflation as governments cannot afford deflation, and hence will over-respond, rather than under-respond.
And what can investors do about it?
For ‘perfect’ protection against the headline inflation, inflation linked or CPI bonds are the lowest risk, best matching asset available for protecting the real value of money over a reasonable term. Given long tenors, credits tend to be super high quality, which further makes them ideal as the base of any portfolio (and much, much better than that great useless metal, gold). The need for inflation protection increases dramatically as one approaches retirement because inflation particularly hurts those with fixed incomes and a fixed capital base. Those lucky enough to be younger hopefully get a lot of inflation protection through wage increases.
The Australian Government has three current bonds (and possibly more soon) issued, with maturities in 2010, 2015 and 2020. There are also bonds issued by NSW, Vic, SA and Queensland. Finally, there are less liquid issues by the banks, property, infrastructure and other Public Private Partnerships, which are often very closely linked to the underlying state government credit. The tenors are typically long (up to 30yrs), which reflects the need for inflation protection over the long term.
What sort of returns do you get from CPI Bonds?
The cost of the lowest risk asset is a low real yield. As a guide, the real yield for the 20/8/2015 Aust. Govt CPI Bond is about 2.50% (+ actual CPI). So if actual inflation averages 5% over the next 6 years, then total return will be approx 7.5%. If actual inflation is 0%, then the return will be 2.50%, etc. Some securities have deflation floors. This 2.50% real yield should be compared to government 2015 nominal bond at circa 4.25% (which gives a break even inflation rate of 1.75% over the next 6 years). Alternatively, it can be compared to cash at 3%. (Note the above examples assume that investors hold the securities to maturity and are wholesale rates).
State Government CPI Bonds tend to have 50-100bps higher real yields. There are a range of other credits available, offering higher yields as the credit worthiness of the issuer decreases. Like all markets at the moment, there are some outstanding opportunities available, if you know where to look…however, there are also some potential disasters waiting to happen.
In Australia, the Index used is the headline Consumer Price Index (CPI) published quarterly by the ABS. The basket measured includes food and oil, but not asset prices directly.
Apart from CPI bonds, what other asset classes to you suggest as useful inflation hedges?
Property is eventually a good hedge. Initially though, one may have to suffer capital mark-to-market losses as, in a very high inflation environment, interest rates will rise. This may lower property prices. So as inflation rises, property prices may initially be falling. Ditto equities in general. Ultimately, however, sustained inflation will increase rent (the earnings) and, over time, will overcome the interest rate effect. But it could take many years (and the benefit probably won’t flow through until the inflation threat has subsided and interest rates are falling).
Equities have been well covered elsewhere - I like LPTs (Listed Property Trusts) and infrastructure companies (eg tollroads) more and more as both leverage and prices reduce. Sydney Airport (as a great shopping centre) has many inflation linked rental streams. The health sector also fits in well (especially if we are hedging longevity too!).
Commodities (such as wheat, corn, coffee, oil and useful metals) should be reasonable ‘real investments’ over time, but with significant volatility and little income. Commodities are generally denominated in USD, so some of their performance is really a short position in USD. Beware.
How would you balance all this up when constructing an inflation-resistant portfolio?
Well, an exact inflation hedge will include only CPI Bonds. This can be achieved through direct investment, or through a managed fund.
However, a higher returning inflation-resistant portfolio may have CPI bonds as the first base investment, but also some direct property, selected equities, low-levered selected LPT/infrastructure, floating rate securities and perhaps commodity exposure. The greater the diversification, the better.
Can a retail investor buy CPI bonds in Australia?
Yes. While the market is predominately institutional ($500k min size, $5m market parcels), many of the above securities are available to retail investors in $1k, $10k or $100k min parcel size. Retail investors can use a retail fixed income broker, such as ABN AMRO Morgans (RBS is a shareholder in this business), RIM securities or FIIG Securities. There is a cost for using these brokers, usually in the form of a lower real yield (higher price) paid. However, they will hold the bonds in safe custody and provide admin over the life of the security. This cost may not be economic for investments smaller than (say) $25k.
For smaller investors, a managed fund may be a more cost effective way to achieve this exposure. I believe UBS Asset Management and CSAM Asset Management offer retail inflation linked bond funds locally. There is also a range of international “real return” funds, with PIMCO the most well known, but many other offered by large global fund managers.
Thank you, Andrew. Those insightful comments are much appreciated.
Custom Search 2
Comments
Thank you for sharing your views with the Brislemouth community, Andrew.
I note Gareth used the term 'true inflation' in his blog post. Do you have any thoughts on whether the CPI is a reliable proxy for genuine inflation?
Might there be political factors at work which provide incentives to understate the real rate of inflation in the CPI calculation?
The Australian Bureau of Statistics (ABS) measures CPI inflation as defined as a basket of goods and services.
Clearly everyone has a different individual consumer "basket" from each other depending on age, lifestyle, and many other factors. I doubt if anyone's individual basket matches the ABS measure exactly. I have faith that the official measure, given the basket chosen, is not maliciously miscalculated.
The basket itself will always be scrutinised at the margin, often by those self-interested. However, there are some flaws. Some overstate, and some understate.
The political factors in Australia probably more influence which measure of inflation (or wages) they index payments/contracts to (or set monetary policy) - rather than the calculation of the measure itself. In less developed countries, there are often stories of the government fiddling the CPI lower (eg Argentina recently).
Ideally individuals have a good idea of what factors may affect their standard of living going forward and adjust their "inflation resilient' portfolio accordingly. Simplifying to make a point, perhaps elderly have a greater weight on healthcare, and agricultural commodities - and perhaps the aspiring homeowner have a greater weight to the property sector.
The CPI is not the perfect measure of inflation, and will almost certainly be "wrong" for individuals. However, Inflation Linked Bonds will almost certainly provide a tighter hedge to True Inflation than any other constructed portfolio for the "Average" consumer at the cost of lower returns.
Andrew, do you think there is any merit to the argument that consumer prices are not, in themselves, a measure of inflation, but rather a consequence of inflation, with inflation better defined as an increase in the supply of money. If this is the case, then measures like CPI may delude us into thinking we're in a low inflation environment for years - as was perhaps the case during the Howard years. I’m no economist, so forgive me if I’m being inaccurate or over simplifying.
I do have sympathy for the monetarist argument. Although, also not being an economist I struggle with how to measure accurately and conceptualise the supply of money in today’s world. What impact dominates, financial sector deleveraging or printing money? So, perhaps my summary is that describing inflation as an increase in the supply of money is theoretically good, but practically difficult to utilise.
With reference to the “Howard years”, I assume you mean that cheap money fed its way into asset prices, and CPI does not directly measure these. The related question is whether central banks should use the CPI as their “inflation target”, or whether it should be a broader measure (which may include asset prices). Central banks have historically found it difficult to identify asset bubbles at the time.
For another opinion, I have asked our economist for his thoughts…below
“ It's a monetary argument, which is that in the long run inflation is a monetary phenomenon.
It relies on the identity that mv=py, where m = money supply, v = velocity of money, p = the economy-wide price level and y = real activity. Since money can't affect real activity in the long term - real activity depends on capital, labour and technology and how efficiently you use them - then a big increase in money should only be reflected in a higher price level over the long term.
All this is pretty sensible, but is extremely hard to apply in practice as innovation has made it hard to measure money, especially when a financial sector is changing (eg, cash withdrawals can be made on credit cards, which are outside the scope of money supply measures). Also, the velocity of money is not stable over time, so you can't reliably predict what a x% rise in the money supply will do to inflation. Additionally, the lag between the money supply and inflation is very, very long (studies usually try to pick relationships over 5/10Y blocks of data).
Central banks were all attracted to the idea of targeting the money supply, but the practical problems involved saw it ditched in the 1970s and 1980s when commodity and labour supply shocks saw inflation spike for reasons other than a surge in the money supply (the RBA ditched it in 1985). The practicalities meant that they shifted to just targeting measured inflation, which is easily observable and has far fewer measurement problems, using interest rates rather than money supply as the instrument of policy because rates are easily observable and more controlled. The transmission mechanism is still imprecise with long and variable lags, but is easier to map out than for targeting the money supply. “
Can floating rate notes, that pay distribution of certain per cent above BBI, be considered a hedge against inflation?
Good discussion, it would be interesting to reverse the problem and look at ways or protecting your assets or profiting in a deflationary environment. It looks like a 50/50 bet whether we will have inflation or deflation.
Cheers
The fact that the effects of the money supply may not be felt for years down the track, and the fact that it is difficult to quantify money supply in the present, is interesting. The cynic in me would say it's dangerously convenient. It would be an incentive for politicians to throw caution to the wind, 'stimulate' their economies today, record rapid economic 'growth' today, be acclaimed as 'the worlds greatest economic managers' - and leave the consequences for tomorrow. Whether it's difficult to measure or not, and regardless of how sophisticated our financial and monetary systems have become, surely we can't escape the basic truth that the Robert Mugaby effect applies to all economies - not just to those economies that we consider ourselves superior to.
If we concentrate just on the benchmark rate (3 month BBSW in Australia), and ignore the credit margin – then floating rate notes (FRNs) certainly provide inflation protection, if you believe the RBA will raise the cash rate in the event of inflation. BBSW is closely linked, and generally a little higher than the cash rate. The RBA is an inflation targeting central bank which provides good comfort of this link. FRNs will definitely do better than many other asset classes, many of which get belted in an unexpectedly high inflationary environment.
To understand further the alternatives, we need to introduce the Fisher equation, where approx. Nominal Yield = Real Yield + Inflation Expectations. In the example higher, 2015 Fixed Nominal yield = 4.25%, Real Yield = 2.50%, and hence market pricing for inflation expectation is 1.75%. One can buy a Fixed Rate Bond and achieve 4.25% if held to maturity. The buyer of the bond is better off if over these years actual inflation is less than expected, and worse off if higher than expected. Alternatively, one could buy a CPI-Linked Bond at 2.50% Fixed Real, and Floating CPI. The owner will achieve a known Real Return over the period, and will be compensated (higher or lower) for actual inflation. Hence a CPI Bond can be thought of part fixed, part floating,
If one buys a floating rate note for this tenor, then you would receive actual BBSW over the term. Currently this rate is close to 3%, and given economic woes, not expected to rise in the short term. If actual inflation surprises higher, then one would expect the RBA to raise rates, increasing BBSW. The risk in buying a FRN is generally in a lower inflation and/or lower real yield environment where returns are reduced (but positive). But clearly, they are compensated in a high inflation, high real yield environment.
Some things to think about:
1. In event of stagflation, the RBA may not be able to increase the cash rate, despite high inflation (due to the low growth). In this case, a floating rate note holder is not being compensated adequately
2. In event of one off increases (eg Oil rise, GST, Carbon tax), then RBA may “look through” the one off impact in setting the cash rate, hence not increasing rates, despite inflation.
3. If the RBA changes the way they conduct monetary policy (ie away from inflatin target), then FRN holders may not get compensated adequately.
I appreciate your work and value your opinion, but i notice that you have a thing against gold. Gold is money and therefore holding gold is a form of savings not an investment. There is nothing wierd about saving money that holds its purchasing power against the ravages of inflation - its worked for 5000 years.
I can see why some might have a small allocation to gold in these times. I’d prefer a long CPI Bond position, and short USD position than gold, as I think it will perform better in 95% of forward looking scenarios. In financial/social Armageddon scenarios (the other 5%), then gold can rally, hard. But so can more commerically useful commodities which might hold up better in other scenarios. There has been some fascinating news recently on China hoarding copper as quasi reserves. Now that’s interesting. If the would muddles thru, I fear Gold will fall sharply.
Hence, I'm not in gold, but many smart investors are, which I respect.
It's said that the serious investor should only buy Government Bonds at issue rather than on secondary markets. But Treasury Notes, Treasury Bonds and Treasury Indexed Bonds are only issued by competitive tender from the Australian Office of Financial Management. How is a competitive tender any different from the secondary market other than simply being the very first sale?
Post new comment