A lively week in US on the data front was somewhat matched by much action in the markets.
The yield curve was perhaps the most interesting change. It flattened somewhat. There have been three or four data surprises over the last few weeks showing potentially higher than expected nominal growth in January: Average Hourly Earnings, nominal retail sales, core CPI and now core PCE. It’s only one month but it seems to fit a pattern.
The US 12 month generic government 12m yield has now risen back up from a 44bps low hit after the poor December business outlook figures were released in January. After Friday’s core PCE data they were at 56bps, beginning to indicate a greater chance of another rate rise this year.
The whole yield curve did not shift by so much. Five and ten year yields have barely budged. This yield curve flattening tells us that the market is not impressed by the short term bumps in inflation and, or, if it were they believed the Fed would tighten to snuff out any such rises. US 5yr breakeven inflation rates have bounced from the <1% inflation lows of the January turmoil.
Equity markets were unsure what to make of the data. The S&P500 was up a little over the week but not without a fair bit of volatility. The February flash survey for the US Services sector, 80% of the economy, was very poor and taken badly, especially on the previous day, somewhat mysteriously. Often, these private sector sentiment surveys seem to be see quite a bit of stock market action in the 24 hours before their official release. The stock market didn’t end the week too well, reacting poorly to the higher than expected Core CPI release – possibly in reaction to the FOMC’s expected reaction. It doesn’t bode well for next week.
The US Dollar also recovered from its swoon, as we expected, indicating tightening monetary policy – in line with rising the short rates and the expected speculation about the reaction of the FOMC to the new data.
Topic of the week: A test of the data driven Fed
Whatever the reason for January inflation numbers, the fact remains NGDP growth is on a very low trajectory. Inflation is a strange concept. Sure the basket of goods and services created by surveys for the CPI and the PCE deflator may be showing a little higher growth, but the really important number remains the non-deflated nominal figures.
It is still extremely difficult to say with any confidence how much RGDP is really rising since separating out the inflation from nominal growth remains so fraught with near impossible tasks.
Changes in the pattern of spending are picked up in the PCE survey, based on actual data, but the deflating remains reliant on hundreds of individual price indexes that are supposed to capture “pure” price effects.
Sure the nominal expenditure figures capture some of the changing mix of spending, as prices forced higher by supply constrains force consumers to switch to other things, or as prices collapse due to demand changes, consumers may take the savings and spend them elsewhere forcing prices up, and to a lesser or greater extent bringing about greater volumes of the same product. Just think about one change we might be seeing in the US at the moment: lower spending on energy being redirected to healthcare, say. But does healthcare increase its production in response or merely jack up prices? So the questions remain, do the indexes, can the indexes, capture the price impacts from these change in demand?
Most mainstream economists, and certainly the Fed and the FOMC, don’t waste too much time worrying about these issues. The Fed claims to be data driven. It was hard to see how the December 2015 rate rise was data driven, it seemed to be more driven by a need to “normalise” rates.
Whatever, the Fed made clear that future moves would be data-driven. Well, now they have a chance to put their money where their mouths are and raise rates. Headline and core CPI and PCE are on clearly rising trends. The 2% target for core PCE could be breached if trends continue. We don’t expect this to happen, but it is merely a typically flaky inflation number, if a bit less flaky than CPI.
The inflation hawks are calling on the Fed to show some leadership. We think it is crazy to tighten monetary policy when nominal growth is both weak and weakening, but then we are not the Fed and not driven by the same data as them.
As well as the PCE price data, the actual PCE data was released too. This showed consumption running at a nominal 4% YoY in January. Real PCE ticked up only slightly to 2.9% YoY.
The Atlanta Fed GDPnow was weak,coming back to 2.1% annualised 1Q16 growth. The inventory element is causing some volatility with the figures.
Inventories in December continued to build to unusually high levels, leading to the upward revision to RGDP in 4Q. The downside is that they are forecast to run down in 1Q16, causing the estimate for 1Q16 in GDPnow to fall back from 2.5% to 2.1%
The underlying Atlanta Fed forecasts for final demand was more steady at 2.5%. The retail trade figure for January earlier in the month had already bumped up estimates for the growth in the PCE contribution to GDP. The PCE data itself bumped up the expected 1Q contribution a bit further, especially as it is for a broader array of purchases. It must be caveated with the fact that there is increased volatility in prices, so separating out the real from the nominal movement in PCE will be even trickier than usual.
The high inventories at the end of 2015 are coming down, meaning less “investment” in inventories and thus less investment spending, cutting the growth forecasts, offsetting the higher PCE. In fact, they may be two sides of the same coin, unexpected goods inventories lead to “clearing sales”, leading to higher PCE “real” spending. The cash saved in the goods sales is then pumped into services, that sees no higher output, just temporarily higher inflation. Who knows? But it is plausible.
The RGDP was revised up a touch, but was still a poor 1.9% YoY, the lowest rate for two years. More importantly, NGDP was still poor YoY at 3.0%, the lowest rate for three years and confirming the new, even weaker, trend begun in 3Q14. If only the Fed looked at this data.
The February Markit Flash Services PMI was very poor, continuing the trend seen in the more official PMI’s ISM Non-Manufacturing (Services) ISM in January.
Industrial production and manufacturing indexes are already actually weak, even if largely energy sector driven. Although service sector employment intentions show solidity, reflected in official job numbers, the business outlook was reported as the weakest since August 2010. Businesses must remain hopeful as they are not cutting costs, which is good, but things are clearly quite fragile.
Our base money watch showed an unusual effect as YoY growth was positive for the second week in a row, but this time the actual amount dropped week on week.
It is mostly surveys coming out in the US next week until Friday when the February official jobs numbers are released. We would expect them to be OK given the hiring intentions of the services sector and the reasonably strong consumption trends seen in January. We do not expect this reasonably happy picture to continue as NGDP is tending too low and base money growth flat at best.
And, of course, the Fed has a tightening bias in place, so any unexpected bad economic news will be taken badly – especially as the inflation hawks are starting to scream.