Below is a summary of my post-CPI tweets. You can (and should!) follow me @inflation_guy or sign up for email updates to my occasional articles here. Investors with interests in this area be sure to stop by Enduring Investments. Plus…buy my book about money and inflation, just published! The title of the book is What’s Wrong with Money? The Biggest Bubble of All; order from Amazon here.
- In prep for CPI: Econs forecasting about 0.15% core; Cleveland Fed’s Nowcast is 0.18%; avg of last 4 months is 0.20%.
- So, econs which have been too bullish on econ for a year (see citi surprise index) are bearish on CPI.
- If we get any m/m core less than 0.20% (even 0.19%), y/y will round to 2.1% b/c dropping off high 2015 April.
- But after that, next 8 months from 2015 were <0.20% so any downtick wouldn’t be start of something new.
- Hard to tell but the core CPI print was SLIGHTLY above expectations. 0.195%, so y/y was 2.147%.
- In other words, if someone charged another nickel for a candy bar somewhere we would have had 2.2% again. <<hyperbole
- That 0.195% m/m was lower than April 2015, but higher than May, June, July, Aug, Sep, Nov, and Dec.
- Core services unch at 3.0%; core goods downticked to -0.5% y/y.
- y/y Medical Care decelerated for second month in a row, down to 2.98% y/y; still looks to be in a broad uptrend from 2% in 2014. [ed note: chart added for clarity]
- Within Medical Care, medicinal drugs accelerated, prof svcs was flat. Hospital svcs dropped from 4.33 to 3.15% y/y
- Hospital services oscillates – we’ll probably get that back to 4%-4.5% which will push med care back up.
- Primary Rents 3.73% from 3.66%. OER 3.15% from 3.12%. Some were expecting deceleration there. Not us!
- Lodging Away from Home dropped to 1.32% from 2.27%. That, and various home furnishings, is why Housing subcat went to 2.12 vs 2.14.
- But Rents and OER are the stable measures…not Lodging, not furnishings.
- Core ex-housing fell to 1.39% from 1.48%, but again that’s due to elements of med care and housing that are likely to rebound.
- Lots of movement within Apparel but overall nothing. The February pop looks like a one-off.
- Overall, a more buoyant number than expected and the stuff holding core CPI down are the transient things.
- Biggest m/m declines: infants’/toddlers’ apparel (-26.5% annualized), fresh fruits & veggies; women’s apparel; Lodging away from home.
- Biggest m/m outliers: Motor Fuel (+152.3% annualized), Fuel Oil, Processed Fruits & Veggies; Motor Vehicle Insurance.
- My estimate of median CPI is actually 0.28% m/m and 2.46% y/y. But…
- …but the median category this month may be affected by regional housing, and I don’t have the BLS factors. So grain of salt needed.
- This summarizes the inflation story. Rents and Services ex-rents both rising ~3%. Core goods is the anchor.
Discussion: after last month’s surprising m/m core CPI print of +0.07%, many were questioning whether that was the outlier, or whether the +0.29% and +0.28% of January and February were the outliers. The answer might be that they are all outliers, as this month’s print was very close to the 4-month average. But even so, +0.2% m/m would produce a 2.4% core inflation number by the year’s end. That’s consistent with what we are being told by Median inflation. Both figures would suggest core PCE, after all of the temporary effects are removed, is essentially at or slightly above the Fed’s 2% target.
There are two pertinent questions at this juncture. The first is whether the Fed will feel any urgency to raise rates more quickly because of this data. The answer to that, I think, is clearly “no.” This Federal Reserve’s reaction function seems to be overly (and overtly) tilted towards growth indicators – and even more than that, their forecast of growth indicators. The majority of the Committee also believes that inflation expectations are “anchored” and so inflation can’t really move higher very quickly. They only pay lip service to inflation concerns, and honestly they aren’t even very good at the lip service.
The second question is where inflation goes next. Whether the Federal Reserve raises the target overnight rate or not, the question of inflation is relevant for markets. And the indicators seem to be fairly clear: the larger and more persistent categories are seeing price increases of around 3% or more, while the main drag comes from a “core goods” component that is highly influenced by the lagged effect of dollar strength (see chart, source Bloomberg).
Recently, the dollar has been weakening marginally but still is in a broad uptrend (looking at the broad, trade-weighted dollar). But if the buck merely goes flat, core goods will start to move higher. And that means even if core services remain steady, core inflation should push towards 3% later this year.
This doesn’t sound like much but it would be highly significant (and surprising) for many observers, investors, and consumers. Core inflation has not been above 3% for two decades (see chart, source Bloomberg).
This means – incredibly – that many students in college today have never seen core inflation above 3%, and more importantly many investors have not seen core inflation above 3% during their investment lives. When core inflation breaches that level, it will feel like hyperinflation to some people! And I do not think markets will like it.
Wise investors hate crowded trades. Good, high-alpha trades tend to be out-of-consensus and uncomfortable. Bad trades tend to be ones that everyone wants to talk about at the cocktail party. Think “Internet bubble.” That doesn’t mean that you can’t make money going along with a consensus trade, at least for a while; what it means is that exiting from a consensus trade can be very difficult if you wait too long, because you have a bunch of people wanting to go the same direction as you.
So what is the most crowded trade? In my mind, it has got to be the bet that inflation will remain low and stable for the foreseeable future.
This is a very crowded trade almost by default. If you want to be long momentum stocks and short value stocks, and no one else is doing it, then it can get crowded but this takes some time to happen. Other investors must elect to put on the factor risk the same way as you do.
But the inflation trade doesn’t work that way. When you are born, you are not born with equity risk. But you are born with inflation exposure. Virtually everyone has inflation risk naturally, unless they actively work to reduce their inflation exposure. So, from the day of your birth, you have a default bet on against inflation. If there is no inflation, you’ll do better than if there is inflation.
It’s a consequence of living in a nominal world. And the popularity of this bet at the moment is a consequence of having “won” that bet for more than three decades. Think for a minute. When you find someone who thinks that inflation is headed higher – and let’s cull from our sample all the nut-jobs who think hyperinflation is imminent – what is “higher” to them? When I tell people that our forecast is for 3% median inflation by year-end, they look at me like I’m from Mars, like three percent is so unfathomably exotic that they can’t imagine it. Because, for the most part, they can’t.
This month marks a full twenty years since the last time that year/year core inflation exceeded 3%. Sophomores in college right now have never seen 3% core inflation. (Median inflation has gotten somewhat higher, up to 3.34% in 2007, but hasn’t been higher than that since 1992). So truly, for many US investors 3% inflation is exotic, and 4% inflation is virtually hyperinflation as far as they are concerned.
So this is a very crowded trade. And this crowded trade is expressed in numerous ways:
- Bonds of course do very poorly in inflationary outcomes. Floating rate bonds do slightly better. Inflation-linked bonds do the best. And inflation-linked bonds, while richer now than they were, are still vastly preferable to nominal bonds in a real risk sense and still quite cheap in an expected-return sense.
- Equities do poorly in inflationary times. While earnings tend to keep up with inflation, the P/E multiple is usually at a maximum when inflation is between 1% and 2% and tends to decline – severely – when inflation moves out of the butter zone.
- While Social Security has a cost-of-living adjustment, very few private pensions do. Some annuities have “inflation adjustment” features, but with very few exceptions these are fixed escalators and not sensitive to inflation at all. There really aren’t any inflation-linked annuities to speak of.
- What about the structure of our workforce? Unions tend to be stronger in inflationary periods because it is during these times that their power (as monopolists in the local labor market) to keep wages moving up with the cost of living is deemed more valuable by potential union members. The chart below is from a presentation I made several years ago.
There are many other examples; most of the ways we express this trade are unconscious since we are so accustomed to living with low inflation that we don’t give our default choice – to face down the possibility of inflation while hedgeless – a second thought. This is, without a doubt, the most crowded trade. And why should investors care? Investors should care for the same reason you want to avoid all crowded trades: when it is time to exit the trade – which in this case means buying commodities, buying inflation-linked bonds, buying other real assets, selling nominal bonds and equities, pressuring futures markets to offer hedging tools (such as CPI futures), borrowing at low fixed rates for long tenors, and so on – investors may find that it is very hard to do at levels they once considered a God-given right. Or in the sizes they want.
Some readers may note that this is the “Most Crowded Trade,” while I just wrote a book about the “Biggest Bubble of All.” Why don’t I just call the “inflation stability trade” the “biggest bubble?” The difference is in emphasis. A crowded trade can be crowded even if it isn’t a bubble (although a bubble also tends to be crowded), and it is no less problematic for not being a bubble. The fact that it is a crowded trade just means that the door to escape is smaller than the crowd that may need to pass through it. In this case, the crowd consists of almost everyone on the planet, aside from the tiny cadre of people who have hedged to some meaningful degree. But it is possible that the trade never has a forced-unwind, a panicky run for the exits. It is possible, although I deem it unlikely, that inflation may stay low and stable forever. And, if it does, then the crowded nature of the trade is no issue. But the trade is indeed crowded.
In the 1970s, investors could be forgiven for not hedging their natural “I was born this way” inflation exposure. There weren’t many ways to do so. One could buy gold, but when the client asked what else the investment manager had done to immunize their outcomes against inflation he could shrug and say “what else could I do?” But this isn’t the case any longer. Investors who want to hedge explicitly against the risk they have implicitly been betting on all along have many options to do so. And, in a low-return world, they don’t even have to give up much in the way of opportunity cost to do so. For now.
So if the crowded trade unwinds…what will managers tell their clients this time?
The Employment report was weak, with jobs coming in below consensus with a downward revision to prior months. It wasn’t abysmally weak, and not enough to change the a priori trajectory of the Fed. If the number had been 125k below expectations or 125k above it, then it may have had implications for the FOMC. But this is a number that has big swings and is revised multiple times. Getting 160k rather than 200k isn’t cause for celebration, but neither is it cause for panic. So whatever the Fed was getting ready to do didn’t change because of this number.
To be sure, no one knows what the Fed was planning to do, so this mainly has implications for the day’s volatility…which is to say that the market quickly went to sleep for the day.
Now, interestingly the Average Hourly Earnings number ticked higher to 2.5%, continuing the post-crisis upswing. At 2.5%, hourly earnings growth is slightly higher than median inflation and thus potentially “supportive of the inflation dynamic” from the standpoint of the Committee. Yes, wages follow inflation but not in the Fed models – so, while I don’t think this has any implications for future inflation it will eventually have implications for Fed policy. But this is a dovish Fed, and 2.5% earnings growth is not going to scare another tightening out of them…unless they were already planning to tighten.
Wages are actually a bit higher than that. Back in April I highlighted the Atlanta Fed’s Wage Growth Tracker and summarized how this measures is better than Hourly Earnings. I hadn’t been aware of this index previously but I follow it now. It stands at 3.2%. The difference between average hourly earnings and the Atlanta Fed Wage Tracker is summarized below (Source: Bloomberg). Again, though: I don’t think we have seen anything today which will change the Fed’s collective opinion about the need for different monetary policy.
Earlier this week, I promised that I would revisit the question of how we can have both deflation and inflation, and how these concepts are confused. I first posted an article summarizing this point in January 2014, and in re-reading it I think it is good enough to pretty much cut-and-paste with only mild edits. So here it is:
How Inflation and Deflation Can Peacefully Coexist
In the discussion about whether the economy is exhibiting “inflationary tendencies” or “deflationary tendencies,” I find that many, many observers grow confused by the fact that we measure prices in dollars, which are themselves subject to changes in relative value due to supply and demand.
It helps to forget about dollars as the unit of measure. Just because it says “One Dollar” does not mean that it is an ever-fixed mark. With apologies to Shakespeare, dollars are not the star to every wandering bark, whose worth’s unknown although its dollar price be taken. There are two ways to look at the “inflation/deflation” debate. Depending on which one you are referring to, deflationary tendencies are not inconsistent with price inflation, and price inflation is not inconsistent with deflationary tendencies.
One is the question of dollar price; and here we are mainly concerned with the supply of dollars and the number of times they are spent, compared to the amount of stuff there is to buy. More dollars chasing the same goods and services imply higher prices. Of course, this is just another way of stating the monetarist equation: P ≡ MV/Q. This is an identity and true by definition. Moreover, it is true in practice: rapid money growth over some moderate length of time always corresponds with rapid deterioration in the purchasing power of the money unit – in other words, inflation. At least, we have no examples of (a) extremely high money growth without high inflation, or (b) extremely high inflation without high money growth.
But this is not the same discussion as saying that “the aging demographic [or debt implosion in a recession] means we will have deflation,” as many economists will have it. Deflation, in that sense, can still happen: if you have fewer workers making the same amount of GDP, then goods (and services) prices will fall relative to wages, which would be deflation the way we typically mean it if the overall price level was otherwise unchanged. However, if the money supply increases by a factor of 10, then nominal prices will increase no matter what else is going on. It may be, though, that in this case wages will increase slightly more than prices, so that there will be “deflation” in the unitless sense.
So, these are not inconsistent statements: (a) there will be increasing inflation next year, and (b) large amounts of private debt and demographic “waves” around the world are a deflationary force. The resolution to the seeming inconsistency is that (b) causes downward pressure on certain prices relative to other prices or, if you ignore the unit of exchange, it causes downward pressure in the ratio of one good that can be exchanged for another. Yet at the same time (a) implies that the overall increase in output in goods and services will be outstripped by the number of dollars spent on them, driving prices higher.
So you should cheer for the “good” sort of deflation. At least, you should cheer for it if you are still earning wages. But do not confuse that concept with the notion that prices in dollar terms will fall. That is wholly different, and unless central banks screw up pretty badly it is not going to happen. Indeed, despite all of the so-called “deflationary tendencies” – most of which I agree are important – I believe prices are going to rise in dollar terms and in fact they are going to rise at increasing rates (higher inflation) over the next few years.
P.S. Don’t forget to buy my book! What’s Wrong with Money: The Biggest Bubble of All. Thanks!
 I kept this sentence…it was true in January 2014, as median inflation moved from 2.06% in Dec 2013 to 2.4% today, but I also believe this to be still true. Only the next leg will probably be faster.
Durable goods orders, ex-transportation, showed a negative print today for the second time in a row. This was expected, in most senses of the word, but while I don’t put too much weight on short-term wiggles in Durables it is hard to ignore the fact that the year/year change in Durables has now been negative for more than a year (see chart, source Bloomberg).
So the weakness in Durables is not new. But it bears noting that the last time core Durables went negative, in August 2012, the Fed followed with QE3 almost immediately. To be sure, at the time core inflation was all the way down at 1.9%, whereas today it is a heady 2.2%…
Look, any Fed watcher right now is and should be confused. Conditions which provoked QE just four years ago are now apparently spurring a tightening bias. Bill Gross can be excused for thinking that the Fed will go back to the old playbook and employ new QE, as he apparently did in his latest Investment Outlook – it is harder to excuse his saying that the Fed should drop money, given that it hasn’t worked yet.
But clearly, something is different in the way the central bank is approaching monetary policy. After all, nothing about this weakness is new. As noted, Durables have been negative on a year-over-year basis for a full year, and the Citi Economic Surprise index shows that economists have managed to be surprised on the negative side for an unprecedented fifteen months in a row (see chart, source Bloomberg).
Okay, the index technically turned positive once or twice for a day or two, but this is still the longest run of persistently optimistic errors that economists have had in a very long time. So this isn’t new – the economy is weak.
Unless…unless what is different is that in 2012, economists were pessimistic (the Citi Economic Surprise index turned positive right before the Fed started QE, which means that either the data was too strong or economists were too negative) whereas today they are more generally optimistic? It would be entirely consistent with how the Fed has been run for the last couple of decades if monetary policy was not being guided by actual data, but by forecasts of data. (See my book for more on monetary policy errors!)
Evidence is pretty clear that recently economists as a whole have not only been wrong, but wrong in a biased way, which is much worse. If you are merely a bad shot, you miss the target in all kinds of directions. But if you persistently miss the target in one direction, then it may well be that your weapon sights are not properly calibrated. The first sort of unbiased miss is not as dangerous, even if too much confidence is placed on the shot, because the errors will even out over time. You might eventually hit the target, by accident. But if the target sights are biased, then you will never hit the target until you realize you’re wrong. You would be tightening when you should be easing. A broken clock is right twice per day, but only if it is stopped and not just systematically two hours off.
Now, regular readers of these columns will understand that I don’t think the Fed should be easing. I don’t think the Fed can fix what ails growth, since monetary policy only affects the price variable, and easy money has only created the conditions for the inflationary upswing we are currently experiencing (Gross also acknowledges this, but sees the inflationary upswing somewhere in the unthreatening future while in fact it is here now). The Fed should have eschewed QE2 and QE3, and have long since begun to drain excess reserves. But what I think the Fed should do and forecasting what I think they will do are two very different things.
I suspect Gross is close to right. Absent some recovery in the real economy – something other than payrolls, which as we know lag – it strikes me as unlikely the Fed will be hiking rates again. Ironically, that may help keep inflation leashed for longer since it will help keep monetary velocity constrained – but I am not confident of that, given how low interest rates already are. Since inflation is very unlikely to wane any time soon, I think we are more likely to see the yield curve steepen from these levels, rather than flatten. A yield curve inversion is not a prerequisite for recession. Inverted yield curves tend to precede recessions only because the Fed is typically slow to lower interest rates in response to obvious weakness. In this case, rates are already low and the Fed isn’t likely to raise them and force a curve inversion. Yield curve inversions are not causal! This next recession may catch some people wrong-footed because they keep waiting for the inversion that never comes.
In my next article, I am going to revisit an issue I first addressed a couple of years ago and which might be especially relevant as recession possibilities increase: the question of how we can have both deflation and inflation, and how these concepts are often confused by those people who are stuck in the nominal world.
Today the 1-year CPI swap rate closed at 1.77%, the highest rate since 2014 (see chart, source Bloomberg).
The CPI swap (which, as an aside, is a better indicator of expected inflation than are breakevens, for technical reasons discussed here for people who truly have insomnia) indicates that headline inflation is expected to be about 1.77% over the next year. That’s nearly double the current headline inflation rate, but well below the Fed’s target of roughly 2.3% on a CPI basis. But at least on appearances, investors seem to be adjusting to the reality that inflation is headed higher.
Unfortunately, appearances can be deceiving. And in this case, they are. The headline inflation rate is of course the combination of core inflation plus food inflation and energy inflation; as a practical matter most of the volatility in the headline rate comes from the volatility endemic in energy markets. I’ve observed before that this leads to unreasonable volatility in long-term inflation expectations, but in short-term inflation expectations it makes perfect sense that they ought to be significantly driven by expectations for energy prices. The market recognizes that energy is the source of inflation volatility over the near-term, which is why the volatility curve for inflation options looks strikingly like the volatility curve for crude oil options and not at all like the volatility curve for LIBOR (see chart, source Enduring Investments).
The shape of the energy futures curves themselves also tell us what amount of energy price change we should include in our estimate of future headline inflation (or, alternatively, what energy price change we can hedge out to arrive at the market’s implied bet on core inflation). I am illustrating this next point with the crude oil futures curve because it doesn’t have the wild oscillations that the gasoline futures curve has, but in practice we use the gasoline futures since that is closer to the actual consumption item that drives the core-headline difference. Here is the contract chart for crude oil (Source: Bloomberg):
So, coarsely, the futures curve implies that crude oil is expected to rise about $4, or about 9%, over the next year. This will add a little bit to core inflation to give us a higher headline rate than the core inflation rate. Obviously, that might not happen, but the point is that it is (coarsely) arbitrageable so we can use this argument to back into what the market’s perception of forward core inflation is.
And the upshot is that even though 1-year CPI swaps are at the highest level since 2014, the implied core inflation rate has been steadily falling. Put another way, the rise in short inflation swaps has been less than the rally in energy would suggest it should have been. The chart below shows both of these series (source: Enduring Investments).
So – while breakevens and inflation swaps have been rallying, in fact this rally is actually weaker than it should have been, given what has been happening in energy markets. Investors, in short, are still irrationally lugubrious about the outlook for price pressures in the US over the next few years. Remember, core CPI right now is 2.2%. How likely is it to decelerate 1.5% or more over the next twelve months?
(**Administrative Note: Get your copy of my new book What’s Wrong with Money: The Biggest Bubble of All! Here is the Amazon link.)
Here is an interesting chart that might suggest flows into oil markets.
Or perhaps it suggests something else. But I put together the chart below (data sourced from Bloomberg) because it seemed to me as if recently crude oil – and other energy markets, but I am focusing on Crude – has been having a nice little spike in the middle of the morning. Often, these have seemed inexplicable to me; other times (like yesterday) there was ostensibly news but it was not immediately obvious that the news was oil-supportive.
This chart shows a composite trading day for June NYMEX Crude Oil futures for the last 20 days. You can see that it appears I am not imagining it: on average, Crude has been rallying about 30 cents per day, beginning about 9:00ET.
I thought this might be flattered by the DOE weekly release, which is at 10:30ET on Wednesdays. So I removed those days and the relationship is still clear (although it makes the rise from 3am to 9am look relatively more important).
I don’t know exactly what this means, but it has “felt” to the old trader in me that the market was tending to gap in that time frame and this seems to confirm the visceral sense. It doesn’t necessarily mean my other gut feeling, that this might represent systematic daily allocations – perhaps to energy or broad commodity ETFs? Or OTC products? Or both? – is correct, however.
I am sure that in my career I have seen weirder reactions, but when the Durable Goods number came out and plainly was significantly weaker-than-expected and energy rallied I must admit it was hard to think of one.
Presumably the reaction was an indirect one: weaker growth means the Fed is less likely to tighten, which means a lower dollar and hence, higher global demand for oil. But that seems a wild overthink. If there is a recession in the offing, and if we care about demand in oil markets (and I think we should), then weak economic growth from one of the world’s largest economies probably ought to be reflected in lower oil prices.
There is part of me that wants to say “maybe investors have finally realized that any given month of Durable Goods Orders is almost meaningless because the volatility of the number makes it hard to reject any particular null hypothesis.” I haven’t done this recently, but many years ago I discovered that a forecast driven by the mechanical rule of -0.5 times last month’s Durables change had a better forecasting record than Street economists. Flip the sign, divide by two. However, I don’t really think the market suddenly got wise.
I also don’t think it’s that a bunch of people got the API report, which came out at 4:30pm to subscribers only, early. That report supposedly showed a large draw in crude inventories when a build was expected – but I’m not into conspiracy theories. That would be too obvious, anyway.
The perspective of analysts on Bloomberg was that the oil market will “rebalance” this year, a point which isn’t very far from the point I made last week in my article “Don’t Forget Oil Demand Elasticity!” But rebalancing doesn’t mean that prices should go higher. They have, after all, already gone quite a bit off the lows. By our proprietary measure, the real price of oil is at a level which implies a 10-year expected real return of about 1% per annum. In other words, it’s within about 10% of fair value quantitatively; given the immense supply overhang that doesn’t seem to promise lots of upside from these levels.
It isn’t just energy. The Bloomberg Commodity Index is 15% off its January lows (see chart, source Bloomberg). Precious metals, industrial metals, softs, grains, and meats are all above their lows of the last 1-2 quarters.
Unfortunately, much of this is simply a function of the dollar’s retracement. The chart below (source: Bloomberg) shows the Bloomberg Commodity Index (left axis, inverted) against the broad trade-weighted dollar. Heck, arguably commodities are still lagging.
But the winds of change do seem to be about. Last week, the Treasury auctioned 5y TIPS; though this isn’t an event in and of itself, the fact that the auction was strongly bid is certainly unusual. The 5y TIPS are a difficult sell. People who are concerned about inflation are typically looking to protect against the long-term. But TIPS also represent real interest rate risk, and if you think inflation is going to be rising near-term then you probably don’t want to own lots of interest rate risk. Sure, you’ll prefer inflation-linked real rates to nominal rates (see my timely comment from January, “No Strategic Reason to Own Nominal Bonds Now”: since then, 10y real yields have fallen 40bps while 10y nominal yields have fallen 7bps), but if you really think inflation is about to rear its ugly head then you don’t want any real or nominal duration but only inflation duration.
And inflation duration has lately been doing really well. The chart below (source: Bloomberg) shows the marked rebound in the 10-year breakeven inflation rate since February 9th.
I think there’s some evidence that the pendulum of complacency on inflation has begun finally to swing back the other way. Core inflation is rising in Europe, the UK, Japan, and the US, and it was inevitable that someone would notice. Ironically, it took energy’s rally to make people notice (and energy, of course, isn’t in core inflation). But what do I care?
If in fact the pendulum of complacency and concern has finally reversed, then both stocks and bonds are in for a rough ride. Bonds may be marginally protected by a dovish Fed, but that only works as long as the inscrutable Fed stays dovish…or inscrutable.
(**Administrative Note: Get your copy of my new book What’s Wrong with Money: The Biggest Bubble of All! Here is the Amazon link.)