Oil prices have certainly recovered since the crises and the industrial production has, without a doubt, increased over the past years.

In this article we will attempt to identify the relationship, if any, between WTI crude oil futures prices and US industrial productivity. It is worth mentioning that monthly data regarding the US manufacturing industry and US Total Vehicle Sales have also been gathered and added to the current analysis. The reason the aforementioned figures have been included into the study is based on the hypotheses that the manufacturing industry is arguably the most crucial industrial sector while car sales should somehow be related to oil demand. This chart will attempt to identify any link among the first three data sets:

Figure_1VitoFeb21.jpg

At a first glance, the relationship among the three data sets (whose time spans all go from January 2013 until December 2013) does not look great. In fact, the spikes in industrial production in March and April and the increased industrial productivity in the last quarter of the year do not seem to influence WTI futures prices at all.

Statistically speaking, the correlation coefficient is extremely low and also negative (-0.3) implying that the link between the aforementioned time series is not reliable, as far as taking trading decisions is concerned. The second set of data that will be analyzed presents the performance of the US Manufacturing Index.

The Manufacturing Index figures are included into the calculation of the Industrial Production data anyway; nevertheless, the great importance of this industrial segment deserves a more in–depth analysis and this is precisely why it is treated as a standalone market. The data, even in this case, do not seem to be well connected to the oscillations of oil prices, with the exception of the June / July / August interval where manufacturing productivity incremented along with WTI futures.

Mathematically speaking, the connection between WTI futures and manufacturing data is positive (+0.22) implying that the performances of this specific segment are more reliable than the performances of the whole industrial sector as far as WTI Crude futures are concerned. Nonetheless, the relationship between the data sets remains rather weak. The oil refining process generates commercial petroleum liquids that are subsequently employed in different industries but a large part of what a refinery extracts from 1 barrel of oil is gasoline, diesel and jet fuel. Hence, it is interesting to understand if an increase or decrease in vehicle sales can, in some way, influences the trend in oil prices.

The next chart attempts to shed some light on this issue:

Fig2VitoFeb21.jpg

The relationship looks more stable. WTI futures dropped until March, spiked in August and plunged again in the October – December period while US total vehicle sales figures oscillated in the exact same way only up to September. Conversely, even though the last quarter was characterized by an aggressive plunge in oil prices, vehicle sales experienced a robust uptrend. Quantitatively speaking, the correlation was as high as +0.53 until September but the yearly correlation for the 2 data sets is only +0.15 because of the last quarter. In fact, the explosion in vehicle sales and the plunge in oil prices were so aggressive that their negative inverse fluctuations managed to significantly lower the overall correlation coefficient from +0.53 to +0.15.

All in all, WTI oil prices, over the 2013, proved to have a stronger connection to the manufacturing segment than to the total industrial sector. Furthermore, the total vehicle sales data had an impressive positive relationship to WTI futures until September but in the last quarter the rapport between the two variables changed dramatically.

Geopolitical factors, such as the Syrian civil war and the Suez channel crises, played a key role in the oil market throughout the entire year. Consequentially, the relationships among WTI futures, industrial production, manufacturing index and total vehicle sales (probably this last one more than all the other data sets) have been heavily influenced by the exogenous variables coming from the Middle East.

= = =

Read another feature story:

Don’t Blame High Frequency Trading For Your Losers