Archive

Archive for May, 2015

Summary of My Post-CPI Tweets

May 22, 2015 2 comments

Here 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.

 

  • CPI Day! Exciting. The y/y for core will “drop off” +0.20% m/m from last yr, so to get core to 1.9% y/y takes +0.29 m/m this yr.
  • Consensus looks for a downtick in core to 1.7% y/y (rounding down) instead of the rounded-up 1.8% (actually 1.754%) last mo.
  • oohoooooo! Core +0.3% m/m. y/y stays at 1.8%. Checking rounding.
  • +0.256% m/m on core, so the 0.3% is mostly shock value. But y/y goes to 1.81%, no round-assist needed.
  • Headline was in line with expectations, -0.2% y/y. Big sigh of relief from dealers holding TIPS inventory left from the auction.
  • Core ex-shelter was +0.24%, biggest rise since Jan 2013. That’s important.
  • This really helps my speaking engagement next mo – a debate between pro & con inflation positions at Global Fixed Income Institute. 🙂
  • More analysis coming. But Excel really hates it when you focus on another program while a big sheet is calculating…
  • It’s still core services doing all the heavy lifting. Core goods was -0.2% y/y (unch) while core services rose to 2.5% y/y.
  • Core services has been 2.4%-2.5% since August.
  • Owners’ Equivalent Rent rose to 2.77% y/y, highest since…well, a long time.
  • Thanks Excel for giving me my data back. As I said, OER was 2.77%, up from 2.69%. Primary rents frll to 3.47% from 3.53%.
  • Housing as a whole went to 2.20% y/y from 1.93%, which is huge. Some of that was household energy but ex-energy shelter was 2.67 vs 2.56
  • Or housing ex-shelter, ex-energy was 1.14% from 0.67%. Seems I am drilling a bit deep but getting housing right is very important.
  • Medical Care +2.91% from 2.46%. Big jump, but mostly repaying the inexplicable dip from Q1. Lot of this is new O’care seasonality.
  • Median is a bit of a wildcard this month. Looks like median category will be OER (South Urban), so it will depend on seasonal adj.
  • But best guess for median has been 0.2% for a while. Underlying inflation is and has been 2.0%-2.4% since 2011.
  • And reminder: it’s median that matters. Core will continue to converge upwards to it, (and I think median will go higher.)
  • None of this changes the Fed. They’re not going to hike rates for a long while. Growth is too weak and that’s all they care about.
  • For all the noise about the dual mandate, the Fed acts as if it only has one mandate: employment (which they can’t do anything about).
  • The next few monthly core figures to drop off are 0.23%, 0.14%, 0.10%, and 0.05%.
  • So, if we keep printing 0.22% on core, on the day of the Sep FOMC meeting core CPI will be 2.2% y/y, putting core PCE basically at tgt.
  • I think this is why FOMC doves have been musing about “symmetrical misses” and letting infl scoot a little higher.
  • US #Inflation mkt pricing: 2015 1.1%;2016 1.8%;then 1.8%, 2.0%, 2.0%, 2.1%, 2.2%, 2.3%, 2.4%, 2.5%, & 2025:2.4%.
  • For the record, that is the highest m/m print in core CPI since January 2008. It hasn’t printed a pure 0.3% or above since 2006.

 

There is no doubt that this is a stronger inflation print than the market expected. Although the 0.3% print was due to rounding (the first such print, though, since January 2013), the month/month core increase hasn’t been above 0.26% since January 2008 and it has been nearly a decade since 0.3% prints weren’t an oddity (see chart, source Bloomberg).

monthlycore

You can think of the CPI as being four roughly-equal pieces: Core goods, Core services ex-rents, Rents, and Food & Energy. Obviously, the first three represent Core CPI. The breakdown (source: BLS and Enduring Investments calculations) is shown below.

threecoreparts

Note that in the tweet-stream, I referred to core services being 2.4%-2.5% since August. With the chart above, you can see that this was because both pieces were pretty flat, but that the tame performance overall of core services was because services outside of rents was declining while rents were rising. But core services ex-rents appear to have flattened out, while housing indicators suggest higher rents are still ahead (Owners’ Equivalent Rent, the bigger piece, went to 2.77%, the highest since January 2008). Core goods, too, look to have flattened out and have probably bottomed.

So the basic story is getting simpler. Housing inflation continues apace, and the moderating effects on consumers’ pocketbooks (one-time medical care effects, e.g., which are now being erased with big premium hikes) are ebbing. This merely puts Core on a course to re-converge with Median. If core inflation were to stop when it got to median, the Fed would be very happy. The chart below (Source: Bloomberg) supports the statement I made above, that median inflation has been between 2% and 2.4% since 2011. Incidentally, the chart is through March, but Median CPI was just released as I type this, at 2.2% y/y again.

median thru march

But that gentle convergence at the Fed target won’t happen. Unless the Federal Reserve acts rapidly and decisively, not to raise rates but to remove excess reserves from the banking system (and indeed, to keep rates and thereby velocity low while doing so, a mean trick indeed), inflation has but one way to go. Up. And there appears little risk that the Fed will act decisively in a hawkish fashion.

Advertisements

Which June Did You Mean, Charles?

Yesterday, Chicago Fed President Charles Evans gave a speech in which he said that he probably leaned towards making the first tightening early next year, as there is “no compelling reason for us to be in a hurry to tighten financial conditions.” The Fed, he said, probably shouldn’t raise rates until there’s a “greater confidence” that inflation one-to-two years ahead will be at or above 2%. This isn’t a surprising view, as Evans is the progenitor of the “Evans Rule” that says rates should stay near zero until unemployment has fallen below 6.5% (it has) or inflation has risen above 2.5%. Yes, those bounds have been walked about; in particular the 6.5% unemployment rate is obviously no longer binding (he sees the “natural rate” as being 5% again). But the very fact that he promoted a rule that set restraints on a mere return to normal policy means that he is a dove, through and through. So, it should not be surprising that he isn’t in a hurry to tighten.

What I found amusing is the sop he threw to the bears. Fed speakers often try to do the “on the one hand, on the other hand” maneuver, but in Evans’ case his heart clearly isn’t in it. He said that “you could imagine a case being made for a rate increase in June.” Notice that he doesn’t say he could imagine a case being made! I am also unclear about which June he means. Does he mean…

(thru Apr) (thru May)
Q1 GDP Q2 GDP Median CPI M2 growth
June 2012? 2.3% 1.6% 2.4% 9.2%
June 2013? 2.7% 1.8% 2.0% 6.6%
June 2014? -2.1% 4.6% 2.2% 7.3%
June 2015? 0.2% 1.0% (e) 2.2% 5.4%

I am not sure exactly what he thinks those darn hawks are looking at, but it seems to me the case for tightening in June is getting worse every year.

Eagle-eyed readers will notice that I didn’t include the Unemployment Rate in the table above. That particular metric has been improving each year, but we know that the labor situation tends to lag the economic situation. The Unemployment Rate is a big political football, but it isn’t particularly useful for policy unless you believe in the concept of a “natural rate” with respect to accelerating unemployment in the overall economy. I don’t: low unemployment tends to increase wages, but has no discernible effect on consumer inflation. Moreover, it appears that the “natural rate” shifts quite a bit over time (6.5% down to 5% in Evans’ formulation, in only a few years’ time), making it look to me like a fairly useless concept.

Yes, of course it makes it more difficult politically to tighten when people are out of work, but since monetary policy is quite useful for affecting prices and not particularly useful for affecting growth, this should be a secondary effect at best. The Fed simply can’t help the unemployed worker, except by holding down inflation for him. In the real world, of course, the Fed Chair is not going to countenance an uptick in rates when unemployment is above 5% or so.

Let me be clear: I think the Fed ought to have tightened in 2012, 2013, or 2014, and they ought to tighten now. I don’t necessarily mean they should guide rates higher, but they should reduce the size of the mountain of reserves via any means a their disposal. But if you are going to argue one year over another year, I think it is hardest to argue that now is the time unless you are merely being guided by the old James Carville adage that the best time to plant a tree was twenty years ago, but the second-best time is right now.

One thing that Evans said that quickens my heart, as an inflation-watcher, is that the Fed “ought to allow” a chance that inflation overshoots 2% that is symmetrical to its chance of falling below it. While he is quintessentially unclear about how he would establish these probabilities – as I have just shown, he seems blissfully unaware that consumer price inflation is already above 2% – the mere fact of treating the costs of inflation misses as symmetrical is dangerous territory. The costs are not symmetrical. The costs of an inflation rate around 0% are very low; some frictions, perhaps, created by wage “stickiness” (even this possibility hasn’t been conclusively established until inflation gets convincingly below zero). The costs of an inflation rate of 4% are much higher, since inflation has historically had long “tails.” That is, once inflation goes up a little, it not infrequently rises a lot. Over the last 100 years, if you take the set of all year-on-year inflation rates above 4%, you find that about one-third of them are also above 10%. This means the costs of a loss of inflation vigilance is must greater than the costs of a loss of deflation vigilance.

Inflation Risks Behind, Beside, and Ahead

To be sure, it looks like growth slowed over the course of the difficult winter. The cause of this malaise doesn’t appear to me to be weather-related, but rather dollar-related; while currency movements don’t have large effects on inflation, they have reasonably significant effects on top-line sales when economies are sufficiently open. It is less clear that we will have similar sequential effects and that growth will be as punk in Q2 as it was in Q1. While I do think that the economy has passed its zenith for this expansion and is at increasing risk for a recession later this year into next, I don’t have much concern that we are slipping into a recession now.

Given how close the Atlanta Fed’s GDPNow tracker was to the actual Q1 GDP figure, the current forecast of that tool of 0.8% for Q2 – which would be especially disappointing following the 0.2% in Q1 that was reported last week – has drawn a lot of attention. However Tom Kenny, a senior economist at ANZ, points out that the indicator tends to start its estimate for the following quarter at something close to the prior quarter’s result, because in the absence of any hard data the best guess is that the prior trend is maintained. I am paraphrasing his remark, published in today’s “Daily Shot” (see the full comment at the end of the column here). It is a good point, and (while I think recession risks are increasing) a good reminder that it is probably too early to jump off a building about US growth.

That being said, it does not help matters that gasoline prices are rising once again. While national gasoline prices are only back up to $2.628 per gallon (see chart, source Bloomberg), that figure compares to an average of roughly $2.31 in Q1 (with a low near $2/gallon).

gasoline

It isn’t clear how much lower gasoline prices helped Q1 growth. Since lower energy prices also caused a fairly dramatic downshift in the energy production sector of the US economy, lower prices may have even been a net drag in the first quarter. Unfortunately, that doesn’t mean that higher gasoline prices now will be a net boost to the second quarter; while energy consumption responds quite rapidly to price changes, energy producers will likely prove to be much more hesitant to turn the taps back on after the serious crunch just experienced. I doubt $0.30/gallon will matter much, but if gasoline prices continue to creep higher then take note.

Inflation traders have certainly taken note of the improvement in gasoline prices, but although inflation swaps have retraced much of what they had lost late last year (see chart of 5y inflation swaps, quoted in basis points, source Bloomberg) expectations for core inflation have not recovered. Stripping out energy, swap quotes for 5-year inflation imply a core rate of around 1.65% compounded – approximately the same as it was in January.

5ycpiswaps

And that brings us to the most interesting chart of all. The chart below (source: Bloomberg) shows the year/year change in the Employment Cost Index (wages), in white, versus median inflation.

eciwagesvsmedian

Repeat to yourself again that wages do not lead inflation; they follow inflation. I would argue this chart shows wages are catching up for the steady inflation over the last couple of years, and for the increased health care costs that are now falling on individuals and families but are not captured terribly well by the CPI. But either way, wages are now rising at a faster rate than prices, which will not make it easy for inflation to sink lower.

Let me also show you another chart from a data release last week. This is the Case-Shiller 20-city composite year/year change. Curiously (maybe), housing prices may be in the process of re-accelerating higher after cooling off a bit last year – although home price inflation as measured by the CS-20 never fell anywhere near to where overall inflation was.

cs20

Inflation risks are clearly now moving into the danger zone. I showed a chart of a lagging inflation indicator (wages), a coincident indicator (energy), and a leading indicator (housing). All three of these are now rising at something faster than the current rate of core inflation. In my view, there is not much chance that core inflation over the next 5 years will average only 1.65%.

%d bloggers like this: