Blain’s Morning Porridge – Dec 16 2022: The Inflation Genie – jumping back in the bottle?
“I’ve met the Emperor of China and the only working British miner, but I’ve never met a ….”
This Morning: Something different this morning – is the inflation threat really about to be corked back in its bottle? My head of research, Ernst Knacke thinks so.. so this morning I’m letting him argue the point on where inflation is headed and how to position for lower inflation next year!
Apologies for intermittent Morning Porridges this week. I was up in London for the whole week, trying to cram a month’s worth of meetings, lunches and social events into four days. I broadly succeeded, but my liver is now officially on holiday. Nothing beats a really good breakfast – which is something I miss when I’m trying to rush out a comment between trains, swimming or early morning calls with clients. London is breakfast capital of the World!
Something a bit different this morning:
I’ve asked my colleague at Shard, Ernst Knacke, our head of research, for his thoughts on inflation. Ernst is an absolutely excellent chap, a South African with the ability to actually look past the headline numbers and consider exactly what they mean. He’s been leading our internal debate on inflation, so I asked him to write a piece for the Porridge – which is below.
Inflation and interest rates dominate where markets are heading. It’s the big part of the big picture. The rest, as they say, is mere detail. Next week my plan for the Porridge is to get back to writing about the detail in markets – looking at issues like just how important the news on fusion power is (important yes, but still nowhere near commerciality), China, UK wages and industrial strife, how big data can beat big companies, and the aviation market (which I’ve been neglecting in recent months, but is fascinating), plus a host of other market fundamentals.
This morning it does look like the market is finally refuting the comment I made last week about even bad news is good news when that’s all market wants to hear. Jay Powell, Fed Head, was bearish and crashed US stocks. The ECB is warning of aggressive hikes to come. In contrast The Bank of England was dovish and crashed sterling. Everything back to normal then!
There is a general understanding around markets the macroeconomic outlook remains complex. There is a perception inflation will ease, but also that it’s not decisively beaten yet, so the interest rate levers remain critical. More rate hikes to break inflation may be required – and that leaves the threat level on a global inflation triggered recession elevated.
How deep and how painful – we don’t yet know, but it will be the big theme in early weeks of 2023. To understand it, we need to think about inflation! Over to Ernst:
The inflation genie – jumping back in the bottle?
by Ernst Knacke, Head of Reseach, Shard Capital
When Inflation gets out of hand, it can wreck economies! Just ask those living in emerging nations whose much more accustomed to living with soaring prices and limited prospects of meeting higher costs. On the other hand, Western European and North American populations have been living well beyond their means on borrowed money, and have not experienced inflation since…forever, really! Suddenly it’s become a shock, unravelling everything consumers, governments and central banks thought they knew and had planned for.
Inflation is a key component into any top-down economic model for investors. Alongside liquidity and economic growth, inflation is a critical variable within any investment framework, and as such we need to understand what’s driving inflation, its longevity, and where it’s going!
There are some factors that lead inflation – like the oil price – and others that lag, happening in response to Consumer Price Index (CPI) when it rises. Higher oil prices lead to higher production costs, manufacturing costs…higher everything! It’s simple to understand why rising oil prices is a leading indicator to inflation. Wages on the other hand, are a lagging indicator – rising in response to higher inflation. If you test wage growth, the correlation is highest with CPI data published about six months earlier.
Our modelling and testing effectively concludes there are two variables with a significant correlation with future inflation, those being:
- i)Energy prices – a combination of natural gas and oil – and
- ii)Money supply, where M2 has the highest statistical significance. M2 broadly consists of cash, cash deposits and savings that can easily be converted into cash.
In forecasting CPI, the next step would be to understand how many months you have to bring these factors forward in order to get the most statistical significance and best correlation. In recent history, natural gas has a high explanatory power, whilst longer term, oil as a standalone variable works better. Due to increasing importance and input into production processes, we use a moving average with a 50/50 split between WTI and Henry Hub as our energy variable. Money supply growth has a longer lag before it shows up in inflation data – anything between 12 and 24 months.
One component of CPI which always stimulates debate amongst economists, is the Owner Equivalent Rent (OER), which makes up the majority of the shelter component in the CPI index. OER is a very sticky component and slow moving due to the way its calculated, however, has a very high correlation with a moving average of aggregate house prices over time – about 10 months forward. On its own, house prices do not have a sufficiently meaningful correlation with CPI to suggest any statistical significance. However, when added as a third variable to an inflation model, the overall model’s statistical significance improves materially.
As a result, our inflation model consists of these three components:
- Money Supply, and
- House Prices
Another important factor, but one which is very difficult to model, is inflation expectations. Inflation expectations are a key driver of sentiment…of consumers, employees, manufacturers…the whole economy!
The problem is the best predictor of inflation expectations seems to be inflation itself, which results in a type of ‘autocorrelation’ which is impossible to model. Inflation expectations feed directly into wage expectations and consequently, wage growth.
Running various regressions, the only statistical significance I found between wage growth and CPI is when you lag wages by about 6 months. Compare this to oil prices, which if you bring oil forward by a month or two, consistently has a strong positive correlation with CPI data. Its evidence suggest wage growth is a lagging indicator with no statistical forecasting ability.
Once we have our variables, the next step was to determine the right weight to allocate to each factor. As you can imagine, this is quite dynamic over time. For example, over a 40-year period, a multi-factor regression would suggest a 40% allocation to energy. Looking at the last 5 years only, money supply was by far the strongest predictor of inflation, and a regression would only suggest 10% allocation to energy. As such, we’ve built a dynamic allocation model, which changes the allocation – within limits – to each variable to ensure the best statistical predictability.
All in all, we know the model will not be 100% accurate – but it will tell us the direction and potential extent by which inflation might change over the next 12 months. We can compare this to inflation expectations data backed out of treasury markets and from surveys from the Federal Reserve. In combination, this will be invaluable in helping decide where the market is likely to go over the coming 3 to 6 months, and what it means to be contrarian!
So where is inflation going?
- Energy prices are down about 30% from their highs earlier in the year. In the absence of a 50% rise, and with a recession expected in 2023, it’s difficult to see year-over-year price changes in positive territory in the next 6 to 9 months.
- Furthermore, following the extreme economic liquidity injection during COVID, we are now seeing base effects playing out as money supply growth fell to its lowest level since the mid-90’s. The final variable is set to decline in the face of rising mortgage rates.
- House prices in the US have started to roll over, and we’re unlikely to see this number grow significantly in the absence of a major decline in interest rates, unlikely to occur until we see labour market weakness.
This leaves us with inflation set to materially collapse in the months ahead. This bodes well for long duration assets! Whilst the ultimate duration play is growth equities, its perhaps still a bit too early to pile back into the Ark ETF. Especially given the probability of a recession and tighter liquidity conditions. However, long duration treasuries look very attractive, especially in the face of positive real yields and the drawdown over the last 18 months.
Ernst Knacke, Shard Capital.
I do hope you may find Ernst’s approach to inflation to be a useful point of consideration. I happen to think he has nailed it… but I would as he’s a mate and a colleague!
Five Things to Read This Morning
Have a great weekend, out of time and back to the day job!