How oil price volatility explains these uncertain times

Guest post by Tom Therramus that is the pen name of a US based Professor. The article was first published on Oil-Price.Net

The numbers say that these should be the best of times for America. The economy has been growing for five years. Unemployment is low. Inflation is almost nonexistent and gas is cheap. Yet, many Americans feel deeply uneasy about their future prospects. Uncertainty is the catchword of the moment. This uncertainty is contributing to growing pessimism and anger – discontent that is no doubt a factor in the unsettled state of the 2016 Presidential race.

If times are good, why do so many Americans believe that it is the worst of times? Have we become a nation of neurotics or is something real going on?

In a 2013 article at, prior to the beginning of the 2014 collapse in oil price, I proposed that turbulence in the oil markets was on the way. In a subsequent essay in early 2015, I suggested that this instability in the oil markets would lead to a period of turmoil in the stock market in the coming year or so.

There was no magic behind correctly foreseeing what unfolded in the oil and stock markets over the last two years. Indeed, the path to connecting the dots started with an observation that many readers have probably already made themselves – namely, that oil prices seemed to have been notably changeable or volatile over the last decade or so.

In the following essay, I describe how I used math tools to dig into oil price volatility – discovering a repeating pattern of instability in global oil markets that has ticked like clockwork for at least the last 15 years. The identification of this rhythm within the volatile price of oil has taken some of the guesswork out of forecasting what may be ahead.

It is my contention that the world has not seen a phenomenon of this type previously and that its emergence marks the rise of a new dynamic with potential to shape our economic and political fate. The uncertainty that many of us feel thus may be far from nebulous, but a shared hunch that history’s engines are shifting gears.

The Wave Pool as a Metaphor for How Oil Prices Cause Uncertainty

The waterpark is family fun on a hot summers day. My kids are now teenagers, and unlike me, they all swim like fish. But when they were younger there was one waterpark attraction that especially unnerved me – the wave pool.

My fear grew from a couple of bad experiences that I’d had early on in this lake-sized stretch of water. Every ten minutes or so the wave pool transforms from relative calm to storm-like turbulence.

During the calm periods I’d sometimes be lured into the pool’s deepest parts. But after the wave engine roiled into motion, on more than one occasion I found myself out-of-my-depth in the churning waters and anxious until I could get back to a safe depth where I was able stand with my head above the surface.

Recently it occurred to me that the wave pool provides a metaphor for a complex, but disquieting phenomena that I’ve have been writing about at for some time now -for previous articles see 2010, 2011, 2013 and 2015.

For anyone who has read one of these pieces you’ll know how I have described that every three to four years volatility in the price of oil surges upwards as if spurred by an unseen wave machine.

Among the tools used to pick out this cycle of oil price turbulence is the mathematical technique of Fast Fourier Transform. When the numbers were crunched by this helpful tool, I discovered that since the year 2000 oil price volatility has been oscillating at a period of 1,024 days. Putting this another way, Fourier transformation revealed that wave-like surges of instability in the oil markets have been erupting once every 3-4 years since the turn of the millennium.

It takes a little squinting – but you can make out this repeating, wave-like pattern on the chart above by eyeballing the wriggling red line spanning the years from 1998 to the present, where a rolling 3-day standard deviation in daily oil price is plotted to provide a measure of price changeability over time.

The tsunami of red spikes centered on 2009 is the most prominent feature of the plot, which I unsubtly arranged to coincide with the wave crashing into tube riders to hammer my point home. However, there are also clusters of large volatility spikes centered on 2001, 2005, 2011 and 2015 – the latter of which corresponds to the collapse in oil price that began in mid-2014.

A further tendency that you might pick out from the chart is that overall volatility is on a rising trend over time.

In sum, what the chart tells us is that for more the last 15 years or so, the level of changeability in oil price has moved generally upwards, with notable bumps punctuating this overall trend every 3 to 4 years.

Oil Price Volatility is Bad News

A few months ago I showed one of my articles to a financially savvy friend. He asked, “What is this – some sort of technical analysis? Are you making any money from it?”
“No!” I said, “You misunderstand. In my opinion, what you are looking at is one of the most serious, yet least appreciated threats to our way of life”.

“If you say so “, was my friend’s reply.

His sardonic response is not unusual. I’ve grown use to family and friends eyes glazing over at my mention of “oil price variation”.

So why is it that I believe we need to pay such close attention to this issue.
Unpredictability in oil price, resulting from rapid changes (up or down) over short time spans, is bad news because oil, and more broadly fossil fuel, is the commodity that is most essential to the operation of a modern economy and the wealth it generates via commerce. The burning of fossil fuel literally powers nearly all work.

Commerce thrives best when the costs of doing business can be anticipated and hedged against. It is harder for companies, entrepreneurs and the average worker to do their important work in the unstable business environment created by unexpected swings in oil price. Lower gas prices may temporarily feel good to consumers, but the broader economy does not do well when the rug is unpredictably pulled from under its feet by a large and unexpected change in the cost of energy.

Oil-Price.Net has Called it Right So Far

One measure of the usefulness of the analysis is its predictive power. As summarized above, so far our forecasts based on analysis of oil price volatility patterns have been impressively on the mark.

Late in 2013, when the oil markets were relatively quiet I wrote that a significant increase in the level of oil price volatility was due. The basis of this prediction was that it was coming up on three years since the last surge in price variance in 2011 – i.e., the cycle length calculated from the Fourier analysis .

On cue, oil price began its epic decline in mid-2014, accounting for the most recent cluster of large volatility spikes on the “wave pool” chart above.

In the wake of the oil price collapse of 2014 – in January 2015 – as the US economy gathered steam and relative tranquility prevailed in global financial markets – I co-wrote an article with Steve Austin suggesting that stock exchange turmoil was a looming prospect.

This prediction was made on the basis of a previously identified relationship in which we had noted that increases in stock market volatility tended to follow oil price variance spikes by six to twelve months.

Again, the forecast proved accurate. Mid-2015 saw a sudden collapse in the stock market of the world’s second largest economy – China. Unusual levels of turmoil in stock markets around the globe accompanied Chinese market volatility, most especially in Asia, but also in US stock indices.

In a US example of this volatile phase, on August 24, 2015, the Dow registered a record drop during intraday trading of over 1000 points.

China and the world’s stock markets have remained twitchy since the mid-2015 spasm. Indeed, there has been further sharp market declines (exceeding 10 %) in China and elsewhere in the opening days of 2016. The next year may see a calmer situation, but it seems likely that during this period stock markets could be more than usually sensitive to black swan-like events.

Black Monday Redux

There was another historical event that we were able to point to in our post of January 2015, which suggested that trouble in the equity markets could be on the cards in the near future. This was the Black Monday Crash of October 1987 – the largest one-day % decline ever in the Dow Jones Industrial average. This unprecedented market tumble followed in the wake of a precipitous 63% fall in oil price, which had occurred in the preceding year (1986) due to in fighting within OPEC.

The Black Monday stock market crash happened nearly 30 years ago, well before the turn of the millenium and the initiation of the distinctive pattern of price variance under discussion here. However, it is related to the present situation in that it may provide a singular example of how a propagating wave of uncertainty sparked by a rapid and unexpected change in oil price may impact and destabilize downstream financial markets.

The Chinese obligingly (from our perspective) also dubbed their mid-2015 stock crash Black Monday. This is an interesting historical echo, given that the turmoil in the stock markets in 1987 and 2015 both followed in the wake of major disruptions in the world’s oil markets.

The financial media took no notice of the curiously repeated instance of the “Black Mondays”. Their horse blinders remained firmly affixed. The conventional ignorance that emerged was that factors internal to China – its economic and political arrangements – wholly accounted for its stock market collapse, with tumult in other equity markets occurring as a knock-on effect.

The Chinese’s not so Puzzling Box

But pointing to China as the basis of all the turmoil, is in our opinion, over simplifying the cause and effect of the situation. Yes, for the past two decade or two, China had more than enough cash juicing around to build innumerable factories, highways, airports, mall and sport arenas. At one point, China was burning as much coal as the rest of the world combined, and gobbling up about half of the world’s aluminium, concrete, steel and copper in use.

That demand, in fact, held up prices and helped countries dependent on exports like Brazil, Australia and Russia prosper. All rosy for go! Could the situation remain so? Of course it could not. We need only leaf back through a few recent pages of US history to see how it played. .

Following the dot-com bubble burst and during the recession of 2001-2002, the Federal Reserve cut short-term interest rates from 6.5 % to just 1% – a historic low. As a result, debt became cheaper, people started buying homes at beat-down prices and the economy picked up pace. However, the (now termed silly) mortgage debts and derivatives pulled the economy down. The Federal Reserve then put interest rates to near zero in 2008 in attempt to boost the economy and stimulate the moribund housing market.

To further speed recovery, the Fed began rescuing banks and pumping money in the form of Quantitative Easing. What is termed Quantitative Easing is nothing but the buying of securities at lower interest rates to pump up the economy. The banks were being handed free money and given free rein to invest it. At the peak of the program, the Federal Reserve was dumping $85 billion a month into the global economy through their bond-buying strategy.

So, for China, the sea looked calm until the Federal Reserve rethought its quantitative easing largesse at $85 billion a month. But when the trillions dried up (not stopped, exactly) the tale changed. China’s debt-financed strategy of building factories and infrastructure stopped working. Demand fell, the growth rate slumped and the cost of raw materials dipped. As we know, the price of oil fell too but the cause is varied and not restricted to the China ‘threat’ factor.

Taking bets that the future will be like the past, over extending oneself on debt premised on the collateral of stably priced and ever growing supply of energy, probably seemed like a good idea at the time. But what happens when oil price starts to wobble on its own clock and it becomes harder to predict the road ahead. China, the US – the driving forces are likely the same – as is the outcome – economic disruption. You see what I am getting at?

A Leviathan Wave Engine

A scientific explanation for what is driving the cycle of wobbles in global oil price- the wave generating engine – remains to be identified. This being said, my view is that it may be a function of the world approaching, or being at the crest of the oil resource-depletion curve – otherwise known as PEAK OIL.

Now – I just wrote a bad word for some- peak oil. OK it’s a phrase, not a word, but before moving on I want emphasize that I’m not saying we’re in danger of running out of oil. In fact, by any objective standard there has never been more of the black stuff around. Moreover, our ongoing ability to pump it up in vast, nearly mindboggling quantities is not likely to change any time soon.

To reiterate, what I am focused on here is the emergence of oscillations in oil price unpredictability as a definite thing over the last few years and a mechanism that might explain it.

As reviewed in earlier essays at, there are a number of theoretical and real world examples of how imbalances in supply and demand at the peak of a resource-depletion curve can give rise to oscillating phenomena.

Predator populations, such as those of lions, often show predictable, whipsawing cycles in relation to the numbers of their prey (e.g., antelope). Studies of predator-prey relationships in the wild hint at the complex math – grounded in chaos theory – that could give some insight into what may be going on with the present cycling volatility of the price of oil.

A predator-prey relationship may be an apt description of the connection between humans and fossil fuels. However, the question arises as to whether the tables have turned, and what now is the biggest threat to whom?

The present consensus is that the main peril from overuse of fossil fuels is global climate change. Serious as this problem is, the time scale over which the consequences of increased atmospheric carbon will play out is expected to be relatively long term. It will be nip and tuck, but my guess is that global climate change will occur at a gradual enough rate that large scale adaption of human populations will be possible.

The more imminent threat, in my opinion, comes from increasingly ferocious instabilities in the world’s energy markets. Largely overlooked, but in plain sight for a least the last decade, the wave engine churns on. Based on the established pattern, another surge of volatility in oil prices is set to break over us in three to four years, i.e., sometime between 2018 and 2020.

It also has to be said, that the cycling 3 to 4 year pattern of the last 15 years may be break down – it is a function of chaos after all. What seems less likely to change going forward is that oil prices will be increasingly unpredictable. Also unlikely to alter in the near future is the widespread assumption that energy future equals energy past – a sentiment that will fuel an ongoing cycle of debt-driven bubbles, which will be subsequently punctured by spikes in oil price volatility.

The automatic leviathan of the wave pool thus seems unlikely to be quieted soon. Until and unless we find a way back to a more comfortable depth, we may be stuck with growing the economic uncertainty and political disruptions that stem from this chaotic phenomenon.

This entry was posted in Energy and tagged , , , , . Bookmark the permalink.

38 Responses to How oil price volatility explains these uncertain times

  1. Willem Post says:


    You have focussed on oil prices.

    Most oil bought, sold, transported, etc., under multi-year contracts with more or less fixed or slowly varying prices.

    Airlines buy only a small part of their fuel on the spot market.

    Utilities buy coal under 20-year contracts, or buy electrical energy under 20-year contracts, with an “smoothing adjuster” for wholesale price variation.

    The volatility in oil prices is a concern, especially for countries selling it, such as Russia, etc.

    Over time, folks find ways to more or less wall themselves off from such volatility.

    Whereas prices do vary, the real-world impacts of these variations are muted, which is not shown by your graph.

    • Euan Mearns says:

      A chart from Tom…..

      • Willem Post says:

        The oil price volatility is much greater than the stock market volatility, and the stock market volatility is much greater than the GNP volatility, all due to the muting effect I mentioned.

        • Hi Willem, Thanks again. I’m not sure what you’re getting at here. For example, it is tough to get at GNP variance with the type of time resolution that is available for oil and stock – so I’m not sure how your conjecture is supported. Perhaps if you could elaborate or could share a chart to illuminate?

          • Willem post says:

            Whereas oil price variations get a lot of attention, their influence on GNP is minor because people make adjustments, such as not taking a long trip, in case of high gas prices, but do something else, and thereby keep most of the economy going just the same.

            I call that the muting effect. Of course, some up and down pricing of oil do affect GNP, but in a minor way percentage wise.

    • Hi Will,
      Thanks for your comment and thoughts. Yes, it is complex and you’ve identified some of the mechanisms by which unexpected changes in oil price might propagate effects into the broader economy. A further tangible example is the fact that much of the investment in tight oil extraction (fracking) in the US is leveraged on debt, bonds and derivatives. As we discussed in a previous post (, major Wall Street banks are estimated to hold around $3.9 trillion worth of commodities contract, the bulk of which are based on oil and were written when oil seemed to be destined to remain above $80 per barrel.

      This being said, perhaps one clear way to appreciate the financial impact of oil price volatility is to look at the graph that Euan was kind enough to post for me below. What it shows is a relationship between an oil price volatility index (red -in this case a 3 day rolling standard deviation of daily WTI oil price) and stock price volatility index – black trace and same rolling standard deviation calculation on the DJIA – through 2006-2016. Three pulses in oil price volatility occurring in 2008, 2011, and 2014 respectively, can be seen upstream of three corresponding periods of notable instability in the Dow.

      As an aside, understanding this relationship has helped with my retirement investments. Following periods of instability in the energy markets I now shift money out of my stock accounts in anticipation of chop in the equity markets in 6 or so months time ; >)

    • oldfossil says:

      Tom Therramus’s article documents the symptoms and invites us to suggest the causes. Willem you have pointed out a mechanism that moderates prices, namely long-term contracts between utilities and producers. What proportion of oil purchasers do the utilities comprise? It might be insignificant. My suggestion would be that certain classes of assets could take about 1000 days to bring into production, because of high capital costs and technical difficulty. Shale could be disrupting the cycle because once prices go volatile, it takes only a few months to commission a well and only a few minutes to decommission it. The lower capital cost of a shale well makes it easier to walk away from.

      • Tx for post oldfossil : ) Interesting thought. Shale, and the ease with which frackers can walk away, may also have a disproportionate knock-on economic effects as well – given the complex of financial vehicles – debt, bonds, junk bonds, derivatives etc – used to underwrite it. As far as banks and Wall Street were concerned, when oil was at $80+ per barrel their investment was backed by tangible assets and considered low-risk. It was almost like printing money. This investment does not too clever now with oil continuing to hover in the $40 range.

  2. Euan Mearns says:

    Tom, an important observation for me is that in 2008 and 2014, the oil price had already declined substantially before volatility exploded. That decline is a symptom of weak demand and / or over supply. In think in both 2008 and 2014 weak demand has played a role. Volatility explodes when sh*t hits fan. Leman in 2008, the technical support being breached in 2014, backed by OPEC complacency. And as you observe this eventually is manifest as crashing markets that eventually awaken to the consequences of economic malaise some months later.

    Unfortunately, by the time volatility hits, it is too late to get out of the oil market, which I have learned to my cost on two occasions.

    Have you a view on the current state and future direction of the oil price? Arthur Berman believes this is a false rally. As does Goldman Sachs.

    • Nigel Wakefield says:

      “Volatility”, as defined by those in the financial world who trade options, is a purely mathematical concept based on day to day fluctuations in the price of the underlying asset. It stands to reason that lower outright prices will result in higher volatility, as day to day prices need to move less to create the same “realised” level of volatility.

      Very simplistically, if crude oil prices move (up or down) by an average of $1 per day at an average underlying price of $33, this would be a higher level of volatility than if they moved the same amount at an average underlying price of, say, $100. In the former case, the daily fluctuation would be 3.33%, in the later only 1%.

      With a regular move of $1 on an underlying price of $33, the realised volatility would be approximately 48%, whereas with a regular move of $1 on an underlying price of $100 the realised volatility would be approximately 16%.

      I’m not sure, however, that volatility, as viewed by those involved in the trading of options, is the same as the volatility being discussed here??

    • Willem Post says:

      Oil demands by GNPs decline somewhat, but producers, for various reasons, appear to be sluggish adjusting to it.

      The reason, in case of the Saudis, is putting the shale folks back into their box.

      As a result over production, oil prices decrease, etc.

      Yet, the EU , and Japan are not growing, need still more QE and even negative interest rates.

      Other debilitating factors must be operative.

    • Tx Euan,
      Based on the data, my conclusion is that a new pattern unpredictability in oil prices emerged around the turn of millenium – and maybe even a few years beforehand. My hypothesis is that this new variance pattern is an emergent phenomenon that results from cycling imbalances in supply and demand that occur at the crest of a resource-depletion curve for commodities like oil. A nice example of such phenomena was provided in a great research paper by Ugo Bardi on the 19th century whaling industry, published a few years back (

      In terms of what’s ahead. If the oscillation holds up, then my guess is that we could see a further surge in spiking oil price variance somewhere between 2018-2020. But who really knows. As I hedge in the piece, the pattern could break down – it is a function of chaos after all. I might add though, if there is another surge of volatility in this time frame, watch out in the stock markets 6-12 months ahead.

  3. Ken Meyercord says:

    Willem – I’m intrigued by your statement “Most oil bought, sold, transported, etc., under multi-year contracts with more or less fixed or slowly varying prices.” It’s something I’ve been trying to get a handle on for some time now as I believe the steep decline in the price of oil since mid-2014 is a result of gaming the market to put the squeeze on Russia. Whether that’s right or not obviously hinges on whether it’s possible to game the price, and to answer that question we need to know how much oil is actually traded on the spot market. I have yet to find anyone who can answer that question. Can you quantify your statement?

    • Nigel Wakefield says:

      While I can’t answer the question of how much physical oil is traded on the spot market (i’d guess a relatively small amount), a lot is traded by reference to futures prices. EG Saudi prices oil at a differential to oil futures; the differential changes reasonably frequently, but in essence the price paid fluctuates according to what futures prices are doing. A lot of physical oil is traded in this way, so to a certain extent it is trading with reference to spot (e.g. Dated Brent) or nearby futures price levels.

      As to gaming the market to “put the squeeze on Russia”, OPEC’s decision not to reduce output in late 2014 was what really put the skids under the oil price leading to the drop from $70 to present levels. My belief is that there were a number of underlying assumptions in this decision; first and foremost amongst them the belief in OPEC that a short term price drop would be sufficient to kill off US shale oil, as marginal costs of production were believed to be $70 or higher at the time.

      Obviously a lower oil price would also have put the squeeze on both Russian and Iranian economies too (both already suffering under sanctions) which would have pleased both the US and the Sunni-denominated OPEC members. I think the US was prepared to sacrifice shale oil for greater geopolitical gain and the benefit that lower oil prices ought to have brought to the US economy….

      In retrospect, this was one of the greater miscalculations of the oil age. Shale oil’s resilience in the face of lower prices has astounded pretty much everyone, though clearly there was considerable stress in the sector with prices in the low $30’s and high $20’s. Many of the shale oil producers have survived simply as a result of having higher priced production hedges in place which have allowed them to continue, though we can see that with LTO drilling rig activity at less than 25% of where it was 18 months ago, there is not a lot of new development going on at current prices.

      The recent rally has brought some relief; allowing producers to look to complete some previously drilled but not fracked wells (some1,800 in total just a couple of months ago) while hedging the production at levels that produce a positive cash flow. This will also allow for debt revolvers to be extended for a couple of years, albeit at relatively punitive interest rates. So the game goes on….

      Higher prices from current levels would no doubt see an increase in drilling rig activity in Permian and Eagle Ford Basins (i.e. completely new production), followed even by new drilling in the Bakken if prices managed to get into the $50’s.

      In other words, the higher we go in the short term, the lower prices will stay in the medium term as shale oil put downward pressure on prices again. OPEC are in a no-win situation. Budgets are based on considerably higher oil prices, so reserves are being drawn down (where available) to fill the gaps. If they reduce output, allowing prices to rally further, the cheapest and quickest oil to market is shale oil…. lowering OPEC’s market share and topping off any rally. OPEC would have to cut massively (my guess about 5 million bpd) to make a difference, and would have to accept many years of lower market share as a result, not to mention the carnage that significantly higher oil prices would create in a global economy that already looks significantly stressed….

      It’s a very interesting and complicated situation. Absent a geopolitical meltdown or a massive, concerted and strictly adhered-to cut by OPEC, I tend to agree with the Goldman Sachs view that we see oil prices fluctuate in a $25 to $45 band while the world chews through US shale oil. How long this might take is anyone’s guess… I don’t subscribe to the “decades” of reserves that some others do, but it could easily be 5, even10, years before the best of the easily accessible, cheap to produce shale oil is depleted.

      • Willem Post says:

        I agree with most of your statements.

        Many shale oil companies are so financially stressed, they need to borrow to get rigs re-started, and no entity would lend them money, until it looks to the lender the oil price stays above about $50.

        • Nigel Wakefield says:

          Hi Willem,

          Agreed that many shale oil companies are financially stressed. Any further lending by banks will be contingent (as it almost always is anyway) on those companies enacting hedges to secure the cash flow to service the debt.

          With the market in contango (forward prices at a premium to spot/nearby prices) there is sufficient value, at present, in the forward premium to enable completion of wells already drilled and possibly also to drill new wells in the more productive parts of Eagle Ford and Permian basins.

          If the hedge value can generate sufficient cash flow to cover opex and service debt with a little left over on top, the banks will lend….

          Better for them to do that than see these companies go into Chapter 11 and throw the “keys” back at the creditors… Banks have no skill set in producing shale oil, and, even worse, would not want to assume the onerous responsibility for post-production cleaning up of well sites. They will kick these problems as far down the road as they can…. if only because it gives them more time to try to offload their portfolios on the unsuspecting….

  4. Pyrrhus says:

    “The economy has been growing for five years. Unemployment is low. Inflation is almost nonexistent and gas is cheap.”

    Well, one of those statements is true, depending on what you mean by “cheap.” As an American with Economics and Law degrees, I am amazed by the gullibility of folks who should know better. Unemployment is in fact extremely high, 23% if measured by the standards used during the Great Depression, and the employment that is out there is heavily part time and very low wage. America has been losing full time jobs steadily for 20 years. The Federal government simply throws anyone who has been unemployed for more than a year out of the workforce…voila, not unemployed!
    Similarly, inflation is low if you don’t count taxes, education, food, energy, and regulatory compliance, probably 60% of the average budget. If you did count them in the inflation number, and therefore the GDP deflator, you would discover that there has been negative growth for most of the last 15 years…
    Amazingly, real incomes for the middle class have been declining for the last 45 years even with the government’s dishonest numbers. But it’s much worse with the real numbers, and this is manifested by declining family formation, extremely low net worth (media $3k) for under 35s, enormous student loan debt and soaring auto debt….

    • Willem Post says:


      Absolutely correct.

      Your comment is a great summary. I suggest you write an article with numbers, graphs, etc.

      I have made a study of some Vermont numbers. Rah-rah, government bureaucrats, and politicians, in Montpelier, were not pleased, as their stories collapsed.

      It is good to have selected paragraphs in reserve for commenting, in case those folks re-erect their stories.

    • Gaznotprom says:

      Yep agreed with P, the numbers are fudged to be polite or more/less blatant lies.

      Cost of living continually rising = inflation low?
      Economy on the up = where? Oil price trundling on the bottom…
      QE = for who?
      Libor not rigged = sure!
      Debt falling = levels reaching beyond 07/08
      Employment up = productivity low
      Wholesale price of electric low = brown/blackouts imminent.

      The market is NOT functioning.

    • Euan Mearns says:

      Pyrrhus, while I hope that Tom will answer this himself, I think you miss the satirical twist to what he has written.

    • Hi Pyrrhus, Tx for commenting. Yes – I guess my point is in some ways the same as yours. Things are not what they seem. Hence, my discombobulated attempt to open the post with a riff on the Dickens – “It was the best of times, it was worst of times…”

      • Euan Mearns says:

        Dang, Tail of Two Cities and not Oliver TWIST 😉

        It was the best of times, it was the worst of times, it was the age of wisdom, it was the age of foolishness, it was the epoch of belief, it was the epoch of incredulity, it was the season of light, it was the season of darkness, it was the spring of hope, it was the winter of despair.

  5. With all due respect to Prof “Therramus” my immediate reaction to his article was that a fast fourier transform was maybe not the ideal way of quantifying oil price volatility. A simpler and arguably more diagnostic metric is the coefficient of variation, which is the standard deviation divided by the mean. Here are the results I get when I use it (the top graph shows the CV calculated over weekly periods and the bottom over monthly periods. WTI prices are daily prices from EIA):

    • Roger, Tx for the beautiful graphs. I also use rolling standard deviation to calculate volatility – e.g., see the graph above posted by Euan for me in response to Willem’s first comment. FFT on these plots is what yields the 1024 trading days estimate of main peak in the frequency domain. I have not done FFT on CoV – but will give it a try.

      • Tom: I saw the chart Euan posted in response to Willem but it plots stock market, not oil price volatility. I don’t know if the two are linked or not.

        I’ve been staring at my graphs – which really aren’t all that beautiful, by the way – waiting for the truth to be revealed to me, but so far I’ve drawn a blank. Take out the volatility spikes that coincide with the the first Gulf War and the 2008 global recession and what you’re left with looks much like noise. But maybe some FFT on the CV values will reveal something I’ve missed.

        • On my plot posted by Euan, the thicker red lines represent the oil price volatility trace – the thinner black trace is stock rolling SD. In terms of variance spikes on your plots, forget about pre-2000, look instead at the clusters of spikes on the top plot 2002/3, 2005/6, 2008, 2011, and 2014. True it is noisy, but I think that these contribute to rhythm that FFT is picking out. IMO it is difficult to view the large and rapid collapse of oil prices in 2014/15 as volatility noise. Also, as I argue in the post, the observation stands in part on its predictive character. Thus far its been pretty on the mark.

          • Hi Roger,
            I asked Euan (post of March 24, 2016 at 7:45 pm) to upload WTI and Brent data calculated as per the 3d rolling standard deviation volatility index that I use. It takes a little squinting at the WTI trace to see the repeating pulses (centered on the hand drawn arrows) of volatility occurring at 3-4 year intervals since the year 2000 – i.e., the rhythm that FFT is presumably picking out. However, in my opinion, the pattern is much clearer when examining the Brent data. Not only can only observe the clusters of spikes occurring around 2001, 2005, 2008, 2011 and 2014 – there is also an incipient pulse that appears to occur earlier in 1996 in the Brent price variance data.

  6. michael says:

    A whole lot of nice grafts signifying nothing. Oil is a tiny number of producers whose price is controlled by a cartel. The speculation is not in the oil buying and selling it is in the tiny market of traders who have heartburn one morning and sell and the next with a full belly buy. All the 1024 day cycle indicates is a herd at work. If you required all traders to pay full price for the oil 99 percent of market volatility would evaporate as the Las Vegas types would be gone. Since the price is controlled by OPEC, putting them on trial for price fixing would most likely send oil back to its true production cost which is about 10 bucks a barrel in a large part of the world.

    • Hi Michael,
      Thanks for your response to the post. As much of my day job involves abstraction of data in graphs and the like – I guess I have a different perspective on how such tools can help understand how things work. Also, unless I have data generated by myself, or sources I trust, I’m nervous about speculating.

      There is a lot unpack to unpack in your comment. Selecting one of your conjectures – what are the sources for the statement that the true cost of oil production is $10 a barrel ? For example, based on what I’m reading, my understanding is that tight oil production in the US breaks even in the $70-80 a barrel range.


  7. Mason Inman says:

    Volatility is certainly important, as the article points out, because of the uncertainty it breeds and the effect that has on planning for the future.

    But I thought this was an important comment (from Nigel Wakefield): “It stands to reason that lower outright prices will result in higher volatility, as day to day prices need to move less to create the same ‘realised’ level of volatility.”

    Putting the same idea another way, people care much more about a 10% change when oil is $100 a barrel than a 10% change when it is $20 a barrel.

    Have any of you tried looking at a different measure of oil price changes, such as % of GDP per person spent on oil, in real terms, to see how that changes over time? My thought was that a measure like that would help show changes in the affordability of oil. You could look at this measure for particular nations or regions, or for the world as a whole.

    Also, since the article is about peak oil, I have to plug my new book, The Oracle of Oil: A Maverick Geologist’s Quest for a Sustainable Future. It’s the first biography of M. King Hubbert, the “father of peak oil,” and is being released by W.W. Norton in April. It’s available from all major booksellers (Amazon, Google Play, Apple iBooks—and for those in the UK, Foyle’s, Waterstones, WH Smith and so on). More info at

    • Hi Mason,
      Thanks for your comment. Enjoyed your article in Nature on fracking and are looking forward to giving your book a read.

      I’m thinking about Nigel’s comment and your suggestion of new metrics for looking at price changeability and can’t say that I’ve got useful responses right away – it’s food for thought nonetheless.

      I would re-iterate all the same that the base measurement is a 3-day standard deviation (SD). Its hard to imagine that the inflationary effects on the dollar cost of oil implicit in Nigel’s comment could undergo levels of variation that were able to cause bias over scales as short as 3 days.

      IMO the power of the index in the context that it is being used comes from the fact that the narrow window sampled rolls and is repeated (1000s of times) at closely spaced intervals over extended periods of years (and decades). Any co-variate impacting and knocking the index off-course would have to be working at multiple and widely spaced scales simultaneously.

      Also worth noting that I’m not using SD here in the traditional statistical sense – but as an index of changeability with time. Putting this another way, I am using SD as a positive definite “derivative-like” number to represent volatility.



Comments are closed.