Something is Clearly Amiss in Manufacturing
As we have pointed out on occasion of last week’s economic data update, we tend to focus on the manufacturing sector for a number of reasons. These are: 1. the manufacturing sector is in fact the largest sector of the economy in terms of gross output; 2. given that the various stages of production of the sector are temporally removed from the consumption stage, it is susceptible to attracting a lot of malinvestment during boom periods (often as a boom progresses, malinvestment will later also flow to the consumer stages as overconsumption surges, but this has only been selectively the case in the current cycle); 3. wealth is generated by savings, investment and production (all of which appear to be in trouble).
Assorted machine tools (adaptors) – typical early to intermediate stage products for use in other production processes
Our friend Michael Pollaro has mailed us two more charts pointing to growing weakness in the sector which we are reproducing below. Meanwhile, another hat tip is due to our friend BC, who has pointed out a recent development to us we were actually unaware of and that we show below as well. As you will see, we can actually observe a few oddities in the current cycle, which were not in evidence in previous cycles.
The first chart depicts the change in national ISM new manufacturing orders compared to the percentage change of the dollar value of core new factory orders according to US census data lagged by 6 months (for all manufacturing and manufacturing ex aircraft). As Michael remarked to us, the ISM data may be skewed by survivor bias in the current cycle.
As can be seen, normally the 6 months lagged change in the dollar value of core factory orders is aligned quite well with the ISM new order index (meaning, the latter is usually leading by about 4 to 6 months), but not this time. Instead, the two series have drifted apart quite a bit. We would argue that apart from survivor bias, there may also be a flaw in the seasonal adjustments to ISM data due to the large swings of the 2007-2009 recession (this is just a hunch).
ISM new orders index vs. change in $ value of core new factory orders incl. and excl. aircraft (census data) lagged by 6 months – click to enlarge.
Of course, the actual dollar value is what these new orders are worth to manufacturers in money terms. The recent plunge has already reached levels last seen during the 2000-2002 recession.
We would guess that there are two main drivers of this development: firstly, the massive slowdown in international trade due to economic weakness overseas combined with a stronger dollar has cut into export orders, and secondly, the plunge in oil prices has put a halt to capex in almost the entire fracking space; the same goes for other commodity producing operations, which are suffering due to the general decline in commodity prices. Given that (largely debt-financed) capex by the energy sector was a major factor in the recovery, we are now witnessing the effect of its ongoing demise.
Sales data from the automotive industry are still looking strong. This industry is a beneficiary of the large domestic market, low interest rates and lower gas prices, but there are two caveats: for one thing, there may be some “channel stuffing” going on, as dealer inventories have been growing for several years. Secondly, car sales seem extremely dependent on the expansion of the sub-prime car loan bubble.
This particular credit bubble is much smaller than the mortgage credit bubble was (with the combined value of owned and securitized car loans amounting to around $1 trillion), but in a sense it is also more dangerous for creditors, as the underlying collateral promises very little by way of recovery in the event of a big surge in defaults (traditionally the LTV of loans for used car is oscillating around 100% and for new cars around 90%).
A slightly dated chart of the growth in sub-prime auto lending, with an appropriate ad – click to enlarge.
Next we show the percentage change (y/y) of the dollar value of new as well as unfilled orders of non-defense capital goods excl. aircraft, as well as the y/y percentage change in inventories of such goods. Here too an emerging downtrend is visible that is consistent with the early stages of a recession, even though it looks not decisive as of yet:
Percentage change (y/y) in the dollar value of new and unfilled orders and inventories of non-defense capital goods excl. aircraft – click to enlarge.
The last time a similar dip came into view, the Fed immediately launched more QE, thereby postponing a downturn and allowing more capital malinvestment to pile up – this time, the Fed is fantasizing about rate hikes.
Credit Conditions Worsen Without a Rise in Interest Rates
The next chart, similar to the first one, shows something very odd. Usually interest rate markets are in an observer-participant feedback loop with Federal Reserve policy. As the Fed seemingly “follows” market rates, some observers, such as e.g. Eugene Fama from the “rational expectations” school, have concluded that it has no control over interest rates at all, but as every market observer well knows, this isn’t entirely true.
What happens is rather that market rates on the short end are driven by growing speculative credit demand, inflation expectations as well as market anticipation of future Fed policy moves. The Fed in turn is indeed influenced by these market signals, which it partly indirectly creates – i.e., it is a feedback loop.
One only has to look at the most recent moves in short term rates, which were first driven up by expectations about a September rate hike, and then plunged after no such rate hike was announced and the useless “dot plot” showed one FOMC member moving his expected federal funds rate dot into negative territory (the culprit not surprisingly was Narayana Havenstein – whose views we have previously discussed here and here).
Moreover, the Fed has actively contributed to the bulk of the approx. 112% increase in the true money supply (TMS-2) between January of 2008 and September 2015. It is very difficult to argue that such a vast addition to the money supply isn’t going to affect gross market rates.
Anyway, at the beginning of a boom, the Fed pumps, either by lowering its policy rates (thus inducing banks to offer credit at lower rates) or as has been the case since 2008, by actively “printing” money in addition to this. When at some later stage price pressures emerge, it attempts to “cool the economy down” by hiking its administered interest rates. Assorted capital malinvestments then tend to be revealed as unprofitable, leading to rising defaults, the liquidation of malinvested capital and the emergence of recessions. This time though, something different is apparently happening:
The annual rate of change of the sum of charge-offs and delinquencies of industrial and commercial loans (black line) compared to the effective federal funds rate (red line) – click to enlarge.
As you can see, there has been a strong correlation between emerging loan defaults and the federal funds rate in the past, as one would expect, as economic busts usually emerge in the wake of an increase in interest rates. Recently, the change rate of the sum of industrial loan charge-offs and delinquencies has however moved into positive territory with the Federal Funds rate firmly stuck close to zero.
This indicates to us that since 2008, boom conditions require regular doses of QE to continue, and that the current rate of monetary inflation (a still hefty 8.35% as of the end of September) may not be sufficient to keep all bubble activities in the economy on artificial life support.
This in turn would jibe with our recent observation that the economy’s pool of real savings is likely in severe trouble. Speaking of savings, we want to point readers to an excellent recent post by Mish on the topic of savings, which inter alia quotes Dr. Frank Shostak extensively, who is a well-known modern-day proponent of the subsistence fund theory (we would also point readers to Richard von Strigl’s book Capital and Production in this context).
Apparently some Keynesian dunderhead once again asserted that people are “saving too much” – ironically on the same day when another headline informed readers that “most Americans have less than $1,000 in savings”. Of course there cannot be such a thing as “too much savings”, and proponents of the so-called “savings glut” theory (such as former Fed chief Bernanke) are confusing numbers piling up in accounts due to massive money printing with savings, which is an error of monumental proportions.
Conclusion
If we didn’t hear every day from well-informed “data-dependent” bureaucrat circles that everything is A-OK, we’d guess that the economy is indeed on the threshold of a bust.
The official message.
Meanwhile, if we are interpreting the implications of recent developments in credit-land correctly, it would indicate that central bank policy is going to continue to remain much looser for much longer than most people currently assume – with all that implies. Primarily it implies a persistent zombification of the economy a la Japan, occasionally interrupted by downturns of steadily increasing severity.
More machine tools hoping for a sale
Addendum: A Slump in Rail Car Orders
The demise of the fracking boom has claimed another victim as Bloomberg recently reported: Rail car orders are in a severe slump, “reminiscent of the early 1980’s” as the author avers. Bloomberg provides this chart:
Rail car orders slump in the wake of the fracking boom’s demise – click to enlarge.
As Bloomberg notes:
“Rail car makers still have plenty of work with a backlog of 122,591 units. That’s down from a record of 142,837 at the end of last year, about five times as many as at the end of 2010 before the crude-by-rail boom began.”
Alas, this is like reporting about a “nice backlog in new housing starts” in 2007. It will simply add to the stock of misguided investments (resp. of goods that will prove difficult to sell once there is an increase in order cancellations).
Charts by: Michael Pollaro, WSJ, St. Louis Federal Reserve Research, Bloomberg
