The primary purpose of tea leaves is to be used for a satisfying cup of pick me up. We do not expect you to challenge that view. However we'd have some explaining to do if we added Tasseography to its raison d'être.

Tasseography is the divination method that interprets patterns in tea leaves, according to its standard Wikipaedia definition.

What does a cup of tea have to do with the price of oil?

The wiki further explains:

“after a cup of tea has been poured, without using a tea strainer, the tea is drunk or poured away. The cup should then be shaken well and any remaining liquid drained off in the saucer. The diviner now looks at the pattern of tea leaves in the cup and allows the imagination to play around [with] the shapes suggested by them. They might look like a letter, a heart shape, or a ring. These shapes are then interpreted intuitively or by means of a fairly standard system of symbolism, such as: snake (enmity or falsehood), spade (good fortune through industry), mountain (journey of hindrance), or house (change, success).”

So it is reasonable to suppose that while most tea leaves, after a spot in the hot pot, do make their way through the strainer into their deathbin, some destined for a higher purpose linger on in the loving gaze of the diviner.

A price chart is much like tea leaves that way.

The primary purpose of a historical price chart of a financial asset is to tell its viewer the past and present of where it's been. Nothing less and nothing more than what's written on the label, not unlike that on a box of fine Darjeeling.

But wait, there is a secondary purpose. Financial analysts trained in the art of technical analysis also look at some price charts and allow their imagination, or more increasingly their computing power, to play around with the patterns suggested by them. These may be suggestive of a trend, mean reversion or something more complex. These patterns can then be interpreted intuitively by means of a fairly standard system of technical charting lexiconography to forecast the future price.

In this way, a parallel of sorts may be drawn between tea leaves and price charts. The comparison becomes more pertinent at this time of the year when financial experts spend much energy attempting to foretell the future of stock markets, interest rates, commodities and currencies in their year ahead outlook reports.

In this article, we pay our own respects to the tradition of year ahead market outlook reports by discussing our view on the likely performance of financial markets in 2016 driven, at least in part, by what we can learn from a particular price chart.

Fig. 1 is the price per barrel of crude oil (WTI) from 1948, adjusted for inflation using the headline CPI and shown in logarithmic scale. The inflation adjustment helps in reducing the distortions caused in price comparisons caused by the changing purchasing power of the currency in which it is denominated. A logarithmic scale makes it easier to compare a gain or fall in the series over different time periods.

Fig. 1: Oil prices (WTI, Inflation adjusted, logarithmic scale)

Oil prices (WTI, Inflation adjusted, logarithmic scale)

Source: Nikko AM

The likely trajectory of oil prices could influence several key outcomes this year such as the likely path of inflation, monetary policy response of global central banks, the value of the US dollar and other currencies, capital expenditure plans of US energy companies and more broadly earnings growth prospects of US corporates, health of US high yield bond market and its knock-on impacts on broader credit markets.

The market price of oil itself is driven by such a large number of economic and macro-political factors that forecasting it is no small task. The range of expert forecasts on oil prices for next year varies from lows in the US$20's to as high as US$ 60 per barrel implying the prospect for oil prices to fall further from their current levels or rise by a huge degree.

Against this wide range of oil price forecasts, we were curious what tasseography of the oil price chart itself would suggest to be the likely outcome. However as readers of the iconic “Hitchhikers guide to the Galaxy” would attest, when it comes to matters of great significance such as the answer to the meaning of life, universe and everything, it is important to ask the right question.

We believe the right question might be to ask where is the likely bottom in oil prices? This is because significant further declines are likely to have a greater impact on broader financial markets than gains.

To give the conclusion upfront: the bottom in oil prices should be in the US$ 30 – US$ 35 range, so we believe oil prices have overshot on the downside.

The Method

Mean reversion is an investment intuition that both fundamental and technical analysts agree to. The average price, shown as the red line on the chart, is the all too famous number of $42.

However, a key requirement for using mean reversion as a forecasting method is that the time series must appear to historically revert to some average value. For instance a Sinusoidal chart is particularly amenable to a mean-reversion based forecast because it oscillates around zero. Chart 1 is clearly not. There seems no centre of gravity to which oil prices converge to with any regularity. Over the last sixty years oil prices have either spiked (in the 70s, late 90's, early naughties) or collapsed (80's, GFC, last couple of years) with such violence that the centre of gravity appears to constantly shift.

However, a deeper dive into the history of oil prices reveals that the largest swings were caused by an identifiable exogenous factor such as Middle East oil shock in 70's, oil glut of the 80's, the China growth miracle of the 90's, and more recently the Chinese slowdown.

Excluding all such periods leave us with the decade post the 80's oil glut and before the China induced take-off in the second half of the 90's. To the extent that current declines in oil prices have much to do with deflating expectations around Chinese growth it seems intuitive to use the average oil price just prior to that growth spurt as a reference.

Over 1986 – 1997, oil prices did indeed seem to mean revert around an average of $34. Factoring in some amount of overshoot hence suggests that a $30 – $ 35 bottom in oil prices is a reasonable expectation for the near term.

The Validation

It's pleasing to note that the US$30 – US$ 35 range stacks up reasonably well against a more educated, fundamentally driven outlook of oil prices.

Oil demand/supply models indicate the continuation of an oversupply situation caused by sluggish demand growth and increasing supply. The politics of OPEC suggest a lack of production discipline will keep supply high even as additional new supply from Iran comes on to global markets. Financial break even costs on the other hand suggests a supply response is more than likely from US shale and other high marginal cost suppliers globally.

Over the past ten years, the commodity supercycle and era of an effective OPEC market “balancing” encouraged entrepreneurs to drill deeper, bigger and pricier wells. Ultra deepwater projects were sanctioned with breakeven costs over $90/bbl, while new sources of oil previously uneconomical were developed such as US shale and Canadian oil sands. Oil producers also ventured into jurisdictions with higher sovereign risk than was previously thought acceptable.

Oil, unlike most other commodities, has a concept called “natural decline”. Unlike a factory producing the same number of widgets year in year out, an oil field left alone will produce less and less with each passing year. According to the International Energy Agency, the average decline is 6.2 percent a year. US shale has an even higher decline rate, estimated at an annual 47% for the Bakken, 55% for the Eagle Ford and 22% for the Permian. These shale plays accounted for a significant proportion of FY15 US production. How do oil producers stop decline? They drill more wells and this requires capital.

Now back to pricing. The halving of the oil price in 2014 took nearly all producers by surprise. In 2015, the attitude of most producers was a “wait and see” approach. Of course, they cut headcount, emphasized the most productive wells, squeezed suppliers and changed project scopes, but at the back of their minds they were waiting to see if this was a temporary blip or a longer term sustained drop. The easy fruit was picked and most producers on average cut costs by 20%. Now was 20% good enough? Probably not - oil prices fell a further 30% in 2015 and into early 2016 they fell a further 20%.

Moving on in 2016, oil producers are facing US$35/bbl and they are facing some testing capital and operational questions in regards to operational decline. The key questions most producers are asking themselves right now are;

  • Can I make an adequate return by investing more capital to stop decline?
  • Do I really want to be an operator in this jurisdiction?

As we move through 2016, most producers will find it a lot easier to say “no” in a US$35/bbl environment than at US$100/bbl and hence we would expect to see the natural wonder of decline occur to help bring the market back into balance.

The steep contango in the forward term structure of oil currently suggests ours is not such a non-consensus call. While spot WTI has conclusively broken below $28 currently, forward prices of March 2017 oil contracts are still trading above $35. Given expectations around market rebalancing in the second half of the year this might seem reasonable.

An answer is still needed, though, regarding why the spot market is still in free fall. We believe this is just because by nature, commodities almost always overshoot. The overshoot may have just become more pronounced this time because of the prolonged period of over-investment caused by artificially low interest rates, elevated geopolitical uncertainty caused by the end of Opec-era monopoly pricing, the potential re-entry of Iranian oil into global markets and the need of US shale operators to keep operating cash flows high enough to evade bankruptcy. Be that as it may, the overshoot will also accelerate the eventual mean reversion as excess capacity is burned off.


In conclusion, we believe oil prices currently have overshot to the downside as long term pricing and current supply/demand dynamics are both suggestive of a logical bottom in the US $30 - $35 range. That's our eureka moment; now for a cup of tea.