Investing During Inflation

Recently I have been looking at whether inflation might be in the pipeline. The jury is still out on that, but caution would be wise given the current situation.

That leads to a rather obvious question: what should investors do during an inflation?

First off, if we are to be naive stock investors, how much does inflation impact stock market returns? We can see the impact in the following chart.

But this chart simply does not capture the pain investors feel during proper inflations. In order to get a sense of this, let’s zoom in on the ‘decade of hell’ – the inflation period 1972-82.

As we can see, basically all of investors’ nominal stock market gains were wiped out during this inflation.

Were there alternatives? Many will point to commodities and their derivatives. I think that this is misleading because, in large part, the 1970s inflation was driven by rising commodities prices. There is no guarantee that inflations will always be accompanied by rising commodities prices.

Okay, so let’s just stick to really basic alternatives. What do they look like? Here are real returns by asset class sorted for various rates of inflation (and including the ‘decade of hell’).

Faced with this, the S&P500 looks like a pretty good bet.

So, if we abstract from alternative asset classes, do we just have to sit content with either low or stagnant stock market returns? Maybe not. There is a chance that we could juice them a little by chasing factors.

The following measures what might be called ‘factor enhancement’. That is, the ratio of the average factor component in a given inflationary regime relative to the total factor average over the whole time series. So, any reading above 1 indicates that the factor is ‘enhanced’ in a given inflationary regime.

Interestingly, even at a first cut, it appears that utilising factors during high inflations may be enough to juice a portfolio. Momentum (Mom) and value stocks (HML) seem to be pretty good bets during inflation; while small caps (SMB) have their day in the sun in very high and moderate inflation.

During the ‘decade of hell’ tilting your portfolio toward the three factors could have substantially eased the pain of low returns.

Update: I have expanded this analysis to the quality factors here.

Posted in Uncategorized | Leave a comment

Inflation, the Quality Factor and Distribution

In a previous post I undertook a very simple analysis to show that factor investing during inflation helps investors to stop from simply treading water – at least, if history is any guide. An interlocutor asked if I’d looked into quality factors and I said that I would get around to it.

In fact, there is something very interesting in the quality factor analysis – something that highlights an aspect of inflation that is not properly appreciated by most economists and investors.

First, let us take a look at how returns stack up during inflations for the standard Fama-French quality factors. Once again, I will be using my ‘factor enhancement’ ratio – that is, the ratio of average returns during various levels of inflation relative to the total average return for that factor. Here are the results.

Note that we have taken seperate deciles, more on that in a moment.

First of all, what is the broad picture? Before getting into it we should note that this data only runs from the late-1960s. So, we only really have access to one inflation. That said, a striking trend emerges. Whereas the investment factor performs unusally well in inflations, the operating profitability factor performs unusally poorly.

This is in contrast to when we tested value, small and momentum, which all saw a boost from inflation. Ditto for utilising quality factors during the 1972-82 inflation: using the investment factor would have been a big winner, while using the profitability factor would have been a big loser.

Why is this? This is where the deciles come in to give us a hint. As the reader can see in the above, the negative impact on the profitability gets worse the smaller the decile used. The numbers above do not quite do justice to this effect, which can bee seen in the chart below.

The inflationary period 1972-82 in this chart is marked in grey. As we can see, the inflation did such damage to excess returns in the 10% decile that this index never really regained its ground after this. Why?

Let’s step back and recall that what this index represents is: outperformance of companies with unusually high profitability. Why would these companies be so negatively impacted by inflation?

Put another way: we know that all of the high profitability companies are impacted negatively by inflation, but as we move up the ladder this impact gets stronger – what does this suggest?

Well, it appears to suggest that inflation has a negative impact on profit margins – and the higher the profit margin, the worse the impact. This brings us to a little known fact about inflation more generally: from a macroeconomic point-of-view, inflation is a distributionary phenomenon.

We can show this using a simple national accounting identity which runs:

This is no abstract economic theorem. It fits the data from the inflation and national accounting statistics perfectly, as can be seen in the chart below (wage share is inverted to make the picture clear).

Recall that in the national account, the wage share is the inverse of the profit share. Translating from national accounting language into investment language: the profit share is simply profit margins. So, we can say that profit margins are ultimately determined by the change in prices relative to the rise in labour compensation and productivity.

We will not go too much into this fascinating identity. Much could be said. For now, we just need to get a sense that profit margins are likely to be impacted by inflation.

All very good in theory, but what did macro-level corporate profits look like in this period?

A mixed bag, really. They did a round trip in the whole period. Needless to say: they were volatile.

That is more than enough for our purpose. It implies that profitability was shaken during this period. It is no surprise then, to see that firms’ chosen for their profitability saw their returns shaken too. With all this redistribution taking place via changes in both relative and aggregate price-levels, it would not be surprising at all to see redistribution of relative profit margins taking place between firms.

The lesson is clear: you probably do not want to rely on a factor during a period when the economic underpinnings of that factor become highly volatile. Hence why the more ‘concentrated’ that factor becomes – i.e. the further up the deciles we go – the more the volaitility of the economic fundamentals impacts the factor.

Posted in Uncategorized | Leave a comment

Podcast on Inflation and Labour Shortages

I appeared on the Bullhouse podcast again to discuss some of the recent work I have done on inflation. I also talk about how the impact of COVID19 and the policy responses to it risk resulting in major labour shortages and a return to a 1970s-style inflationary regime.

Podcast: Inflation and Labour Shortages

Posted in Uncategorized | Leave a comment

Forecasting Future Inflation Using Private Sector Indices

In our last post we saw that private sector indices for used car prices and for rent prices were highly predictive of future changes in the corresponding CPI component indices.

The next logical step is obvious: we should use this information to build an aggregate CPI index that factors in this forward-looking information to get a prediction of inflation over the next six months.

In all honesty, I was a little reticent to do this. Not to put too fine a point on it, but it is a pain in the rear to do this sort of forecasting. You need to strip the CPI series right down using the component weights and then rebuild it using the private sector forecasts and weighting them accordingly.

But, since inflation is such a hot topic right now, I figured it was probably worth doing this as I have not seen it done elsewhere. So, here are the findings. (See last post for forecast methodology).

First of all, here are the component forecasts using lagged Zillow Rent Price data and lagged Manheim Used Car price data.

As we can see, based on the lagged private sector indices, future rent price increases should accelerate and future used car prices increases should decelerate.

Now, here is what happens if we weight these indices and plug them back into a reconstructed CPI index.

What this tells us is that the accelerating rent prices largely offset the decelerating used car prices. This means that we should not expect CPI inflation to decelerate in the coming six months, but we should not expect it to accelerate either.

Of course, this forecast is based on forward-looking private sector indices for rents and used cars only. I have not tried to model the impact of rising house prices on the index – so-called ‘Owners’ Equivalent Rent’. I may do so in the future, but I would need to find a forward-llooking proxy for that too – probably from Zillow. I’ll see how much attention this series gets.

Posted in Uncategorized | Leave a comment

Using Private Sector Data to Forecast Future CPI Moves

Back in early August, my old colleague James Montier and I released a White Paper on inflation. In it we argued that our baseline scenario was that we would see transitory inflation caused by the extreme supply shocks caused by the lockdowns. We drew an analogy to the end of rationing in Britain after WWII, where we saw temporary price increases in the markets for rationed goods.

Further, we argued that any sustained inflation would require a wage-price spiral. That is, in order for inflation to feed on itself, the price rises would have to give rise to wage rises that then spurred further price rises.

Since we wrote this, it appears that there are, in fact, anomalies in the labour market. Many service sector jobs appear to be finding it hard to get workers. Some are blaming this on unemployment benefits, but it seems far more likely to me that many workers are simply too frightened to return to work. It is well-established that the average person vastly overestimates the danger of the virus and so it is no surprise that a certain portion of the workforce are not willing to work. What is more, the ‘fear effect’ is even more pronounced in young people who typically work service jobs.

Anyway, we are not going to deal here with the potential for sustained, structural inflation. Needless to say, the risk is real. And the more transitory inflation we see, the greater the chance that a sustained inflation will take hold.

Private sector metrics of sectoral inflation have far outperformed public data in terms of predictive power. One of the core drivers of the uptick in inflation we saw in the second quarter of 2021 was used car prices. Yet if we were examining the Manheim Used Vehicle Index rather than the CPI we would have seen this two months in advance, as the chart below shows.

David Goldman – whose columns are a must read for anyone trying to get out in front of inflationary trends – spotted this in real-time.

Today we are seeing a new trend emerge – one that Goldman has again been ahead of the pack on: rent price inflation. If we look at the Zillow Observed Rent Index (ZORI) we are seeing rent prices rise sharply.

We can use this information to forecast future increases in the CPI Rent Primary Residence Index. First we need to figure out how far back the CPI rent index lags the ZORI. If we lag ZORI by 6 months we get a very nice fit – yielding an RSQ of 0.6454.

We can then use this to produce a forecast of the future CPI rent index.

So, how much of an impact will this have on the CPI index as a whole? Well, consider this: ‘Used Cars and Trucks’ makes up 2.75% of the total CPI index. And a sharp uptick in used car prices was able to have a major impact on the overall index.

‘Rent of Primary Residence’ makes up a full 7.86% of the total index – almost three times as much as used cars. That said, we should not expect rent prices to increase as sharply as used car prices and our forecast suggests that they will not. But this is still cause for concern.

Posted in Uncategorized | Leave a comment

Podcast Appearence on the Bullhouse

Yesterday I appeared on the Bullhouse podcast with Kenna for a very interesting discussion about a wide range of financial and economic topics.

We discussed a number of different topics. Some of these were as follows:

  • Whether there is a property bubble in the US and internationally right now.
  • The potential bubble in the junk bond market and its implications.
  • Why the risks in the market for rental properties are idiosyncratic this time around.
  • How real estate is now a financial asset and why it should be thought of in context with a person’s broader portfolio.
  • The specifics of the London property market and the idiosyncratic shock we saw there last year.
  • The debt-for-equity swap and how low interest rates have not led to the ‘euthanasia of the rentier’ but rather has led to an evaporation of equity – both stock equity and now home equity – for the middle classes.
  • Why the private equity move into home markets is probably going to end up being a bad investment.

Real Estate Market with Philip Pilkington

Posted in Uncategorized | Leave a comment

Systematic Detection of Housing Bubble – Part II: Linear Trends and Points-Based Detection

In the last piece on detecting housing market bubbles, I an through some of the problems with using standard-scoring. Here I will provide two solutions; one complex, the other simple. But before we move forward with this, we must understand one other problem when it comes to determining whether a housing bubble is indeed a housing bubble.

This problem is not, like standard-scoring, related to how we manipulate the data. Rather it is related to how we define a housing bubble itself. The best way to approach this is to give examples. Below I show two very different real house price series; one from Norway, the other from Ireland.

Here we see two series with two very different properties. The Norway series shows steady, upward growth in real house prices over the past 30 years. Whereas, the Irish series shows a boom, a bust and another boom.

Are we implying that if a series behaves like Norway then it is not a bubble? Not necessarily. It could easily be a bubble that has not burst yet – after all, our a priori assumption should be that house price growth should largely track inflation and so real house prices should be largely stable. But there could be reasons why real house prices rise in a certain country that is not due to a bubble.

Norway is an excellent example of this, because it has large oil reserves. These reserves render the population more and more wealthy through time, even if they do not increase their productive capacities. This could, in theory, result in ever increasing real house prices. Does this mean that Norway is definitely not experiencing a housing bubble over the past 30 years? No. But it provides a reasonable explanation for why they might not be.

Seeing a stable linear trend in real house prices over a long time horizon then, reduces our confidence that this trend is the result of a bubble. We can be fairly confident that rising Irish home prices are bubbly because we’ve seen this movie before in Ireland – hence the jagged data series. But when it comes to a country like Norway we cannot be nearly as sure.

The reader will probably have noticed that I included R-squared statistics in each chart. This is because we can use this statistic to get a sense of how tightly linear a series is. If we see an R-squared of 0.8 or over in a country’s real house price index, this makes us less confidence that rising real home prices are a bubble. This gives us one variable for our model.

But now we have to solve the problem with standard-scoring we explored last time. We can do this by taking our base from a point at which the series was stable. Typically when standard-scoring we use the mean and the variance of the entire series. But if this is throwing the results off, when can choose a more stable base. Unfortunately, this will give us statistics that are impossible to interpret in terms of actual standard deviation, but they will work much better when trying to detect worringly elevated real house prices.

Combing through the series, I found that the most universally stable base was between 1993 and 1996. This was after the Japanese and Swedish bubbles had deflated and before the infamous 2008 bubbles started to inflate in the late-1990s.

We can then assign a points system to determine whether there is a housing bubble in a given country. Because elevated prices are more important than linear trend, we will give this priority. Therefore we will assign two possible points on the basis of elevated prices. One point is assigned if the Z-score is higher than 20; the other if it is higher than 35. An additional point is assigned if the R-squared of the series in below 80 – this gives a point for a lack of linearity.

Here are the results (full results in the appendix).

IRL3
USA3
ZAF3
AUS2
CAN2
DEU2
EA172
ESP2
FIN2
FRA2
GBR2
SWE2
BEL1
CHE1
ITA1
JPN1
KOR1
NLD1
NOR1
PRT1
DNK0

Is there a simple number that tells us of a definite bubble? No. But if we accept our premises then the higher the score, the more likely that country is to be in a bubble.

So, the $64m question: are the world’s property markets bubbly right now? Based on our model we would say: yes. Of the 21 countries/regions under scrutiny, 12 have achieved a score of 2 or over – with 3 achieving a perfect score of 3. Only 9, by contrast, had 1 or less – and only 1 had a score of 0.

Finally, we would note a much simpler way to detect bubbles: by eyeballing the time series and making a judgement on whether or not it looks like a bubble or not. In the vein, we present the time-tested and much underappreciated practice of… just looking closely at panel data. (Varible codes below panels).

1 = ‘AUS’
2 = ‘BEL’
3 = ‘CAN’
4 = ‘DNK’
5 = ‘FIN’
6 = ‘FRA’
7 = ‘DEU’
8 = ‘IRL’
9 = ‘ITA’
10 = ‘JPN’
11 = ‘KOR’
12 = ‘NLD’
13 = ‘NZL’
14 = ‘NOR’
15 = ‘PRT’
16 = ‘ESP’
17 = ‘SWE’
18 = ‘CHE’
19 = ‘GBR’
20 = ‘USA’
21 = ‘COL’
22 = ‘ZAF’
23 = ‘EA’
24 = ‘EA17’
25 = ‘OECD’

Appendix

Full results.

Zscr-93-96BaseRSQZ>20Z>35RSQ<80Bubble Score
AUS150.90.951102
BEL32.20.941001
CAN53.80.891102
CHE12.20.460011
DEU21.30.011012
DNK19.60.800000
EA1728.00.711012
ESP34.30.321012
FIN26.70.741012
FRA30.20.801012
GBR141.10.841102
IRL38.80.431113
ITA-0.40.000011
JPN-10.10.720011
KOR-0.30.090011
NLD16.20.600011
NOR28.60.981001
PRT6.60.160011
SWE165.30.931102
USA88.20.611113
ZAF56.60.701113
Posted in Uncategorized | 1 Comment

Systematic Detection of Housing Bubble – Part I: Z-Scoring and It’s Problems

In a recent piece in Newsweek that got some attention, I made the case that the United States is currently experiencing a housing bubble. The next logical question is obvious: are other countries? After all, the 2008 meltdown was a global crisis; the US was not alone in its housing bubble.

In order to try to detect housing bubbles ideally we would like some sort of systematic framework that we can deploy. The problems with using this approach when it comes to hosuing bubbles, however, are not widely appreciated. In this post I am going to highlight some of the core problems here and in the next post a will propose a remedy.

The most obvious approach to building a systematic housing bubble detection is to get access to a widely published time series for multiple countries and then subject it to standard-scoring (Z-scoring). Standard-scoring works by telling us how unusual a datapoint is relative to the other datapoints in the time series.

So long as we are comfortable with manipulating large datasets, this should be fairly straightfoward to do. OECD publish solid databases on a variety of housing price metrics for most of the developed world countries. This data mostly goes back to the 1990s, so the sample should – in theory – large enough to subject to standard-scoring. We will use real house prices as they tend to use the two most internationally comparable datasets (house prices and CPI).

Here is what the housing markets for major developed countries look like when we apply this technique.

At first glance this looks good: if my argument in the Newsweek article is correct, then we know that the US is experiencing a housing bubble and that seems to show up here as a 2.5 standard deviation event. But to take this at face value would be enormously misleading. We can see why if we take some of these time series raw and plot them.

Let us look briefly at three seperate countries with very different housing market dynamics. We will take Germany, which has not experienced a bubble and a crash since our data begins; Ireland, which experienced possibly the largest housing bubble in history in 2006-08; and the US, which had a large housing bubble in 2006-08 and which we suspect to be in a bubble again.

Let us comment country-by-country and compare these more intuitive readings with the more abstract standard-scoring readings above.

Germany’s housing market looks somewhat overvalued relative to history. But it might be a stretch calling it a full-on bubble. Yet if we turn back to the standard-score, Germany looked like the worst bubble in our entire dataset – the country’s housing market is experiecing a 3-standard deviation event (something that has a less than 0.13% chance of happening!). The reason that this is happening should be obvious to those who understand the mathematics of standard-scoring: since Germany’s housing market has been so ‘boring’ for the past 30 years, even a secular uptrend in prices will stand out as a highly unusual event.

Ireland’s housing market went nuts in the last cycle. We can see the ‘biggest bubble in history’ clearly: real house prices increased around 300% from their 1990s levels! Yet if we look at where Ireland is today it appears that they are once again in a bubble – albeit one not quite as bad as last time. Yet if we turn back to our standard-scores, Ireland is near the bottom of the list with only a 1.2 standard deviation event. Once again, if we understand the mathematics we will readily know what is going on: because Ireland has had such a wild ride in the past, datapoints that should register as highly unusual do not.

The United States is somewhere in the middle. We see the housing bubble clearly in the data. It is, as I wrote in the Newsweek piece, as bad as the last bubble. And lo and behold, this register perfectly well on our standard-scoring framework as a 2.5 standard deviation event.

These examples allow us to draw a general inference about using standard-scoring for housing bubble detection: standard-scoring only really works when the volatility – i.e. the standard deviation – of the datasets under examination are relatively similar.

But, to paraphrase Tolstoy, since all housing markets are bubbly in their own specific way the standard-scoring framework – applied naively – fails completely when trying to systematically detect bubbles in various housing markets. Lessons to be learned here about deploying standard-scoring generally, I should think!

Posted in Uncategorized | Leave a comment

Quantifying the Impact of Vaccine Failure on Earnings Per Share

In a post last week, I raised the possibility that the vaccines might not get the virus under control this winter. Since the markets still seem to be pricing in vaccine success, this could have implications for investors. How might we think this through in more depth?

One way to do this is by looking at the Google Mobility Index and seeing if it is any good at explaining EPS growth in the S&P. Here I take an aggregate construction from the index that encompasses all of the economic variables – transport, retail, grocery and workplace movement. I also push the mobility index forward one quarter which seems to give a more reasonable fit – presumeably due to an accounting lag with the EPS data.

Looks pretty good. Here is the same data in linear regression space.

Okay, so at least histroically there appears to be a strong relationship here. It also makes logical sense as the mobility index should be tracking levels of footfall related to economic activity.

Because the EPS index always comes accompanied by handy analyst forecasts this means that we can pull out an implied mobility index forecast – assuming the linear relationship shown in the above regression. We can also construct a scenario where restrictions are reimposed (details in the appendix). Here is what they look like.

This tells us exactly what we assumed in the last post: namely, that EPS forecasts – and presumeably therefore price action in the stock market – is implying a slight improvement or at least stability in footfall related to economic activity; that is, no more lockdowns or similar interventions. By contrast our – very conservative (see appendix) – lockdown-simuluation shows a slight deterioration.

Using our alternative mobility index we can then pull out an EPS forecast and compare it to the analyst forecast.

As we would expect we see that EPS goes back into decline. Not a very marked decline, mind you – we have made a very conservative assumption for lockdown severity. But enough of a reversal that it would likely capture the attention of the markets.

So, is this model realistic? Maybe, maybe not. It is perfectly possible that businesses are now perfectly well adapted to lockdown and lockdown-lite interventions and that the correlation between the mobility index and EPS will breakdown this winter. But we are making very conservative assumptions about the impact of the lockdown on mobility for this reason.

I reckon that this is not an unreasonable estimate. Anyway, the real point of this exercise is to put some numbers on a much more important intuition. An intuition that leads us to ask the question once more: are markets ready for vaccine failure and possible lockdown-ish interventions this winter?

Finally, let us see what EPS looks like if we assume a severe lockdown scenario. This is a scenario where the lockdown is as punitive as it was in autumn-winter 2020 and we assume that businesses and consumers have not adapted since then. I do not believe that this is a reasonable estimate – for one, it seems unlikely that many red states would lock down even if blue states did this year – but it is a nice exercise to work through to see a true worst-case scenario.

Ouch.

Appendix

I estimate the impact of a lockdown as such. I take the percentage decline from Q2 2020 to Q3 2020 and apply it to Q2 2021/Q3 2021. I do the same for Q4.

This is conservative because I am taking percentage decline instead of actual decline. We can see how this is conservative if we take an example.

Between Q2 2020 and Q3 2020 we saw a 12% decline in the Google Mobility Index. Applying this to the same period in 2021, the index declines from -44.6 to -50.2. If we took the actual decline between Q2 2020 and Q3 2020 (-9.6) and applied it to the 2021 data we would see the index decline from -44.6 to -54.3. This effect would be amplified further in Q4 which would fall to -62.2 rather than -54.8 using the percentage decline method.

As I have stated above, I use the percentage decline method to make a conservative assumption when considering future locdown-ish policies because there has probably been some level of economic adaption since last year.

Posted in Uncategorized | Leave a comment

Double Bubble Trouble

Two weeks ago I wrote a piece for Newsweek outlining potential troubles in the junk bond market. I pointed out that there is a strong possibility that enormous junk bond issuance is floating companies that otherwise would have gone bankrupt due to the lockdown measures. Here is that piece:

The Next Financial Crisis is Coming

But that is not the only bubble on the horizon. The lockdowns and work-from-home appears to have driven investors pretty kooky because we also have what appears to be a major housing bubble inflating. I have outlined this in a piece I published today which can be read here:

Are We About to Repeat the 2008 Housing Crisis?

If you add up the employment in the threatened sectors you get a range of anywhere between 8% and 10% of total employment in the United States. In contrast, during the 2008 crisis – which was almost wholly driven by a housing bubble – only around 5% of employment was under direct threat.

It is hard to come to any conclusion other than that, if I am right about the bubbles, the economy could be under more threat from a financial crisis-cum-deep recession than at any time since the Great Depression.

When you make a claim as large as this and you’re not a permadoomer, it’s usually good to ask the question: what would it take for me to be wrong? So far as I can tell we would need to assume the following for my thesis to be incorrect.

  • The enormous increase in debt issuance by companies with balance sheets destroyed by the lockdowns highlighted by the BIS paper I cite is completely sustainable.
  • Revenues are going to soar for these companies in the coming months and they will pay down all the excess debt.
  • Further, the BIS stress test model – which is quite conservative and does not even assume another lockdown – would have to be totally wrong.
  • With respect to the junk bond market itself, the current very narrow spreads we see – especially relative to forecast default rates – would have to be a permanent feature of reality; presumeably this would be due to some permanent Greenspan put-style arrangement implicitly promised by the Fed.
  • Implicit in the last point is that the Fed can actually control junk bond spreads, even during a market meltdown or crisis.
  • With respect to the housing market we would have to assume that the screaming valuations – which are just as high as in 2008 – are now a permanent feature of reality. These will either continue to expand indefinitely – rendering houses more and more expensive – or it would stabilise at its new high-level.
  • The record levels of private sector residential investment growth is either sustainable or will not end with a bang but rather draw down slowly over time as we replenish the nation’s housing stock – the 2008-era lingo for this, now much derided, was ‘soft-landing’.
  • MBS spreads, artificially lowered by the Fed buying up around 30% of the market, will remain suppressed allowing for the current levels of mortgage lending to continue; the record rates of growth of MBS issuance will either continue on or experience a soft-landing.

I think those are the assumptions you have to make to think that I am totally wrong about these bubbles. If you find them unreasonable – I do – then you have to conclude that we could be sailing into seriously choppy waters.

Posted in Uncategorized | Leave a comment