The Great Inventory Growth Seesaw

Includes: DIA, IWM, QQQ, SPY
by: The Motley Monetarist


Inventory variables have always been used to predict future growth, especially the ISM new orders relative to inventories in manufacturing.

Two different inventory variables were considered to predict growth trends in 2016.

It was found that the level of new orders to manufacturing in the last quarter of 2015 would predict a slowdown of growth in the latter part of 2016.

The inventory growth seesaw

"Hangman, hangman. Hold it a little while. I see my friends. I think I see my friends. Coming riding a mile." Led Zeppelin.

Two conflicting reports on the state of the US economy came out the past week. On the one hand, housing starts fell by 3.8% in December. Building permits were revised downwards by a tad. (But as we all know, a lot of the way markets interpret statistics is based on emotion, so any revision downwards, even if small, will be interpreted the wrong way). On the other hand, factory output rose by .5% in January and industrial output by .9%

How to interpret these divergent views on the American economy from the housing sector on the one hand, and manufacturing on the other. One possibility is to look at the hidden mass behind all of these phenomena: namely inventories. More specifically, inventory build-up.

Inventories are an important component of total GDP. Whether finished goods inventories, "pipeline" inventories, inventories can be used to predict future trends in GDP. The following provides three different inventory measures, and to what extent they lead future trends in growth.

See-sawing down the ISM new orders inventory survey.

A traditional predictor has been the ISM (Institute of Supply Management) monthly survey. This report is issued based on a survey of 400 industrial companies in different industries. Two variables released in the survey are new orders and current inventories (the PMI variable) relative to the ISM survey over the past two years.

New orders track profits. Inventories behave cyclically. Declining inventories indicate that production will generally increase in the next few months. Increasing inventories generally means overproduction. In short, the difference between new orders and inventories in manufacturing might be a lead indicator of future growth, with a negative value indicative of a decline in future growth.

The graph below shows that over the last two years, the variable new orders minus inventories only started to lead GDP in the third quarter 2015. The table that follows provides the correlations between the inventory variable and future GDP growth at different leads. One can see that at least if manufacturing is concerned, the inventory variable provides a good predictor for growth 3 quarters ahead. (It should be noted that I have examined these correlations over the past two years of data-results may be different if the period is extended).


Lead one quarter on GDP growth

Lead two quarters on GDP growth

Lead three quarters on GDP growth

Correlations between new orders-inventories in manufacturing and GDP growth





Source: Institute of Supply Management and Federal Reserve Bank of St. Louis

Production smoothing and the inventory sales ratio.

As an alternative to the variation between new orders and inventory, one might consider the inventory sales ratio as a variable to predict future GDP growth. The reason this variable is important is that if firms tend to smooth production, they might hold inventory against future sales especially if future sales are volatile. This might also be used to deal with seasonal variation in sales. Firms could draw down inventories in the fourth quarter when sales are known to be higher to minimize costs of increasing production.

It is not abnormal to have GDP more volatile than sales, and definitely not the inventory sales ratio. For instance, if firms enjoy increasing returns to production, they may prefer to produce more for a certain number of weeks and then un-employ their workers for a lesser stint. On the whole, this might reduce costs, if returns to scale are increasing. Indeed, over the last two years, this seems to be confirmed in the following graph. The table that follows provides the analogous predictability of the inventory to sales ratio.

What is amazing is that this variable has weaker predictive powers (at least if one considers the last two years) than the ISM new orders to inventory variable in manufacturing. Given the declining role of manufacturing in America, the results below are pretty amazing, both in terms of the low predictability of inventory to sales globally as well as the weaker relative predictive power of this variable compared to manufacturing new orders.


Lead of one quarter on GDP growth

Lead of two quarters on GDP growth

Lead of three quarters on GDP growth

Correlations between inventory to sales ratio and future GDP growth





Source: Federal Reserve Bank of St. Louis.

So what do inventories really say?

At least if we consider the past two years, it's pretty clear that inventory build-up relative to new orders for the manufacturing sector presages a pretty pessimistic consensus on growth in 2016. If the past is to be considered a good indicator, growth is scheduled to decline in the latter part of 2016, given the decline in the last quarter of 2015.

Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.

I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.