No one should have been surprised by Putin’s actions regarding Ukraine after the Olympics. While Americans may have a short memory, that is generally not said about Russians. Putin would have learned from Brezhnev’s experience that it is better to invade a country after you host the Olympics, not before. He might also have learned to pick your country and limit your goals. He has been pretty successful so far.
However, there was a range of behaviors that would have been a very strong predictor for the tough Russian actions this past weekend–market behavior, or MARKINT. Financial markets are strong indicators of where people think prices are going. I co-authored an op-ed in the NYTimes (with Jim Rickards) which ran on 21 December 2008 that suggested that plumbing market data for intelligence related information would be extremely useful. This was before the wave of interest in using “big data” for predictive purposes. Omnis Inc, (a consulting firm I am associated with) has been developing this notion extensively. Christina Ray who boasts a strong “quant” background in the financial community has cross fertilized this experience with intelligence questions. If you happened to be watching markets and in particular certain segments of the markets, you would have seen some strange things last Friday. I copy an analysis by Ray below. Suffice to say that if you believe that information travels among senior Russian politicians and investors (sometimes they are the same person) then you might expect them to take positions in the market knowing in advance of actions that will move markets over the weekend.
Preliminary Analysis of Ukrainian Events
As of 3 March 2014 10:30 EST
Christina I. Ray
Omnis, Inc.
cray@omnisinc.com
Introduction
The events in the Ukraine have had a substantial impact on global markets over the last few trading days, but have especially affected markets this morning (3 March 2014). The nature and magnitude of absolute and relative market movements are particularly meaningful for three reasons.
First, consistent with the principals of Market Intelligence (MARKINT), behavior both prior and posterior to the weekend events may provide some predictive indicators of Russian intent.
Second, more specific market behavior (e.g., the performance of Russian Aerospace and Defense stocks relative to their global peers) might provide some insight into whether substantial military action is anticipated by those either privy to such information or with expertise on this subject.
And third and perhaps most importantly in this case, since the US’s and others’ response to the Russian actions is primarily economic and financial (e.g., freezing assets, embargos), the magnitude of expected effect on market may very well be an input into Russian decision-making on next steps.
Please note that these analyses are only rough and preliminary, but can serve as a quick recap of market consensus opinion.
Effect on Global and Russian Markets
The effect on Russian markets is very significant. Figure 1 below compares the relative behavior of a number of major global indices to that of the Russian MICEX stock market index. (The time series show intraday price behavior at a 3-minute frequency).The price behavior for each time series is normalized to a percentage of its initial price three trading days ago (i.e., 27 Feb 2014).
As shown in Figure 1, the Russian MICEX index is down to 86.84% of its value just three trading days ago, or a very substantial decline of 13.16%. Conversely, although all the major indices are down, the magnitudes of the declines are far less, ranging from a decline of 3.26% for the German DAX index, 1.80% for the UK FTSE index, 2.14% for the Japanese Nikkei 225 Index, and only 0.18% for the US S&P 500 index (although the one-day changes are something greater).
Moreover, as the patterns over time indicate, although the non-Russian indices exhibited roughly similar behavior over time (during those periods when their trading hours overlapped), this was not true for the Russian index. It is typical for most major indices to exhibit somewhat correlated behavior, because in general systemic economic conditions drive all stocks. When one or more indices exhibit anomalous behavior relative to their global peers, this may be indicative of idiosyncratic factors affecting the country or region.
This appears to be the case over the last few days. Note the time period just before 08:00 on 28 February. While the non-Russian indices are trading higher, the Russian MICEX index is trading lower, and generally losing ground relative to its global peers. This may reflect the influence of informed trading: either inside information about Russian intent over the weekend, or expert opinion by those familiar with likely outcomes and who are willing to make market bets to profit from their insight.
Figure 1: Comparison of Major Stock Market Indices to Russian MICEX Index
In general, many investors execute a so-called “flight to safety” when events of such international significance are occurring. The US dollar is one such haven, as is the Japanese yen and gold. Given that the US is more embroiled in this particular event than is Japan, an examination of Japanese market behavior may be particularly illuminating.
Figure 2 specifically exhibits the relative behavior of the Japanese Nikkei 225 index to the Russian MICEX index over the last seven trading days.(Please note that the straight connecting lines connect periods where trading hours do not overlap). Over this time period longer than that of Figure 1 (and one which included earlier, provocative events), the MICEX declined 14.32% while the Nikkei actually advanced 1.80%. And, much of that difference occurred over the weekend.
Figure 2: Comparison of Japanese NIKKEI and Russian MICEX Stock Market Indices
Effect on Foreign Exchange Markets
We can also examine the behavior of the foreign exchange markets, again focusing on the relationship between the yen and the ruble. In Figure 3 below, the top graph shows the yen/ruble cross rate over the last two trading days. (The blue “N” markers are showing important news events as included in Bloomberg’s archived news.)
Take particular note of the spike just before 14:57 on 28 Feb. There is an anomalously rapid rise in the cross rate (indicating weakness in the ruble relative to the yen). This timing is significant, in that many futures markets close in the US at 15:00 hours, or just three minutes later. This may indicate new positions taken in a hurry in expectation of weekend events. (Certainly anyone who made such a bet at this time made a great deal of money by Monday morning).
Also note the cross rates of the yen and the ruble relative to the US dollar, as shown in the middle graph of Figure 3. The yen strengthened 0.7% relative to the US dollar over the two-day time period, while the ruble weakened by 1.41%[1].
And as shown in the bottom graph of Figure 3, two general global threat metrics –the price of spot gold (XAU) and the price of crude oil futures (CLA), appreciated significantly over the same time period, rising 2.42% and 1.41% respectively.
Figure 3: Behavior of Japanese Yen and Russian Rubble Exchange Rates and Other Threat Metrics
Stock and Sector-Specific Behavior
Finally, we might examine sector-specific behaviors. Figure 4 below shows a comparison of a custom index of the four largest Russian Aerospace and Defense contractors relative to the broad MICEX index over the last 10 trading days[2].
Note that there is very little difference in the relative behavior of A&D companies versus those in the broad index; although the A&D stocks are down a bit less than the broad index, the magnitude is not particularly significant, and may indicate that there is little expectation of large-scale military action in the coming months. In other words, this may be a case of the dog that didn’t bark in the night.
Figure 4: Behavior of Russian Aerospace and Defense Contractors Relative to Broad MICEX Index
And finally, we can specifically examine those companies most likely to be affected by the Ukrainian events relative to their peers. News stories indicated that Chevron is one of the companies whose revenue is most likely to be affected by the events in the Crimean, as shown in one news story published by Bloomberg news as an update on 3 March 10:53.
Figure 5: New Story (BN 3 March 2014 10:53:47)
Consistent with this story, Figure 6 below illustrates the behavior of Chevron (CVX) relative to its peer set of energy companies included in the S&P 500 index (X5ENRS) for the last three trading days.
The top graph shows the individual behavior of Chevron and the energy index (again, normalized to 100 at the start of the period), and illustrates the rapid intra-day decline in Chevron’s price relative to its peers. (This methodology normalizes for common drivers of all energy companies, and helps to identify the idiosyncratic drivers of Chevron). The middle graph shows the ratio of the two prices.
Naturally, other companies and commodities that can be identified to be particularly sensitive to events in the Ukraine might be similarly examined.
Figure 5: Behavior of Russian Aerospace and Defense Contractors Relative to Broad MICEX Index
Conclusions
Monitoring changing market prices may well provide some indications of Russian intent. Metrics created from market prices may provide either predictive indications of interest, and concurrent metrics whose magnitude may provide indications of changing strategic intent. And as previously mentioned, there may very well be a feedback loop here: Russian actions affect market prices, and market prices affect subsequent Russian responses.
This last point is somewhat bolstered by the timing of Russia’s actions. The timing of its actions (late Friday night in the US, and after markets closed in Europe and Asia) may be some indication that Russia wants to mitigate the effect of the invasion on its stock and foreign exchange markets. Therefore, one might conjecture that the huge impact on market prices being observed this morning is a consensus view by market experts that Russia’s actions may seriously impact its economy.
[1] Note that a lower value indicates a stronger currency in this graph; for example, one gets fewer yen and more rubles per US dollar at the end of the time period.
[2] Some of the large spikes in the A&D index may be due to price reporting errors, which greatly influence the price of an index with only four members.