Get Indexology® Blog updates via email.

In This List

VIX Hit Below 10 in Q2: What Do We Know From the Past?

Tips and Tricks in Reading the Persistence Scorecard

Mightier than Le Pen

Inflation and the Fed

Asian Fixed Income: The Birth of Bond Connect

VIX Hit Below 10 in Q2: What Do We Know From the Past?

Contributor Image
Berlinda Liu

Former Director, Multi-Asset Indices

S&P Dow Jones Indices

With the French election ending in the defeat of Le Pen, one more risk factor has been removed from the table and low volatility has returned.  In Q2, the CBOE Volatility Index (VIX®) fell below 10 seven times, and the closing level of 9.75 on June 2, 2017, was the lowest since 1993.  Do these low VIX levels predict a rising stock market?  Is it time to long VIX derivatives?

VIX measures the expected volatility of the S&P 500® over the next 30 days using S&P 500 (SPX) options.  Since stocks tend to fall much faster than they rise, a high VIX level, or a high implied volatility of SPX options, tends to indicate falling stock prices.  However, the opposite does not always hold true in a low volatility environment, as stocks may either rise or fall.  Low VIX levels only indicate that the market is quiet and that people are not willing to pay a lot for short-dated protection.

Although VIX is just one year older than Justin Bieber and we don’t have enough data points to draw any conclusion that is statistically significant, we can still take a look at the 16 days during which VIX went below 10.  Our observations are:

  • Falling VIX does not necessarily lead to rising equity markets;
  • VIX spot tends to rise due to its mean reverting property; and
  • With the VIX futures curve in contango, market participants can incur losses if they have a long position in VIX futures when VIX levels are extremely low.

Exhibit 1 shows the 16 days in history during which VIX hit below 10, along with the subsequent returns of the S&P 500 over near-, mid-, and long-term periods following the drops.  The immediate performance of the equity market is mixed; the S&P 500 had about a 50/50 chance of going up and down—in other words, it’s a random coin toss.  The 1-, 3-, 6-, and 12-month returns are also not convincing for a strong bull market.  In fact, we noticed that the 9.89 reading on Jan. 24, 2007, was followed by the 2008 financial crisis only ~18 months later.  Overall, one can argue that low VIX levels have told us virtually nothing about stock market returns in the past.

Low VIX levels are usually followed by a positive VIX return, because VIX usually fluctuates around its local mean (see Exhibit 2).  As long as sub-10 does not become the new normal, VIX will most likely pull itself back into the double-digit region.

However, a low VIX level is not a sign of taking a long position in VIX derivatives.

Given that VIX is not tradable, market participants might consider using derivatives such as VIX futures to access volatility.  In the past, when VIX had hit below 10, VIX futures were always in contango, with the futures contracts more expensive than the spot.  That means market participants with a long position in VIX futures incurred losses.  Imagine a market participant holding a long futures position when VIX goes under 10.  Upon the expiration of the current futures, he/she may choose either to let the futures expire or roll to the next contract.  In the former case, the futures price drops as it converges to the spot; in the latter case, he/she has to pay the price difference in roll to the next futures contract.

Exhibit 3 shows the same 16 days during which VIX hit below 10, along with the subsequent returns of the S&P 500 VIX Short-Term Futures Index, which simulates a hypothetical rolling position from first-month VIX futures to second-month futures to maintain a long VIX futures of one-month constant maturity.  Because VIX futures were not available in 1993 and 1994, we do not report the hypothetical returns.  For all of the other 11 occurrences, the futures curves were all in contango.  Although one-week performance was mixed, in the subsequent month following VIX dropping below 10, the S&P 500 VIX Short-Term Futures Index posted negative returns.

Volatility in the U.S. equities market has dissipated, as market participants brushed aside unknowns from healthcare bills, tax reform, and financial regulation.  Low VIX numbers witnessed in Q2 indicate that financial markets may be settling into a low volatility region.  However, history has shown that low VIX levels are not necessarily followed by a rising market in the near future.  More importantly, although VIX spot tends to rise due to its mean reverting property, volatility investors should consider the deep contango associated with low VIX spot, which, contrary to the “buy the dip” principle, may suggest selling VIX derivatives rather than buying.

The posts on this blog are opinions, not advice. Please read our Disclaimers.

Tips and Tricks in Reading the Persistence Scorecard

Contributor Image
Ryan Poirier

Former Senior Analyst, Global Research & Design

S&P Dow Jones Indices

Much has been written about the persistence of manager performance.  The S&P Persistence Scorecard attempts to track the status of top quartile and top half managers over specified subsequent periods.  Over the years, we have received a fair amount of inquiries from our readers regarding the computation, methodology, and interpretation of results.  This blog attempts to address some of the recurring questions and explore some unique features of the scorecard.

The first table in the scorecard shows the headline “performance persistence” figures.  The table uses non-overlapping, one-year periods and is reported over consecutive three- and five-year periods, showing data for the top quartile and top half.  During each one-year period, all of the funds are ranked by the trailing one-year return.  They are then labeled as either being “top quartile/half” or “not included in top.”  The base year, against which all subsequent years are calculated, comprises the funds that we labeled as being in the top quartile over the earliest one-year period.

Exhibit 1 shows this number to be 568 for All Domestic Funds.  Reading the table left to right, we see that 568 * 34.86 / 100 = 198 funds were in both the top quartile for the base year and the subsequent year.  Furthermore, we report that 568 * 1.94 / 100 = 11 funds were in the top quartile in all three one-year periods.

Exhibit 2 shows the same performance persistence over five consecutive periods.  It is important to illustrate this longer time horizon for two reasons.  First, it is generally harder than random chance to remain in the top quartile the further out in time we study.  Second, the most recent results display a particular intricacy; namely, as of March 2017, two funds persisted in All Domestic Funds while no funds accomplished such a feat in the sub asset level categories.  The question then arises: shouldn’t the number of funds on the asset class level at the end of the period aggregate to the All Domestic Funds category?  The answer is no for reasons we will outline below.

The only additive part of the performance persistence is that the sub asset class category counts should add up to the All Domestic Fund category for the base year.  All subsequent years are subject to an interaction effect that makes them not directly comparable.  For example, suppose that all of the unique large-cap funds return 20%-30% each year.  Next, suppose that all of the unique mid-, small-, and multi-cap funds return 0%-5% each year.  Within the large-cap category, it could be the case that no single fund is in the top quartile each consecutive year (displayed in Exhibit 2 as 0% as of March 2017).

A similar result is seen for the remaining asset class categories.  Now, when we combine the universes to examine All Domestic, there will be 100% persistence.  This is because in our simple example, all of the large-cap funds have a return strictly greater than the rest of the universe.  This implies that the top quartile will be approximately the large-cap universe year after year, thus leading to persistence even though the individual categories had none.

The final table format is a transition matrix.  This tracks the movement of funds between quartile/half buckets through time.  From the latest results, as shown in Exhibit 3, we see that of the 528 funds that were in the top quartile, only 31.82% of them managed to repeat that performance over the subsequent non-overlapping period.  Similarly, 21.02% of the bottom bucket disappeared and 8.14% changed their mandate.

With this blog, we attempt to lay out the key steps in interpreting the scorecard so that the results can lead to meaningful educational conversations about manager performance.  These charts, along with the full S&P Persistence Scorecard display the difficulty that fund managers face in trying to repeat their success consistently.

The posts on this blog are opinions, not advice. Please read our Disclaimers.

Mightier than Le Pen

Contributor Image
Hamish Preston

Head of U.S. Equities

S&P Dow Jones Indices

We publish a series of dashboards at the end of each month to summarize market performance in a number of regions. These dashboards include monthly reports on Risk and Volatility, as well as Factor Dashboards for the U.S. and for Europe on a quarterly basis. With June behind us, a half-yearly summary is in order. Highlighted below are two key themes and trends that have emerged in Europe during the first two quarters of 2017.

The French election caused concern, but the Centrist was mightier than Le Pen.

There were significant elections in Europe during the past six months. Those in the Netherlands and (especially) in France were seen as barometers for Euroscepticism on the continent. They also allowed investors to see whether there would be a continuation of the trend towards economic nationalism and populism that has been observed in various countries globally.

Following the surprise results in several votes in 2016, many investors had come to treat polling data with some suspicion. The possibility of ‘Frexit’ in the event of a victory for Marine Le Pen’s National Front heightened the tension. This nervousness over the outcome of the elections and focus on political risk can be observed in the implied volatility of the currency markets. Taking the average of three such measures – the CBOE/CME FX Yen Volatility IndexSM, the CBOE/CME FX British Pound Volatility IndexSM, and the CBOE EuroCurrency Volatility Index – it is clear that investors’ volatility expectations for the three currency markets have trended downwards over the past year. Nevertheless, there was an increase in average volatility expectations for the three currencies leading up to the French election and a fall once the outcome was observed. We could also observe this pattern around the time of the U.K. Brexit referendum and the U.S. general election.

Brexit weighed on equity market performance

After months of speculation about when the U.K. would formally notify the European Union of its intention to leave, Prime Minister Theresa May announced that the two year negotiation period would commence at the end of March (point A in the chart below). Rather than turning attention to the negotiation immediately, however, the Prime Minister’s decision to call a snap election (point B in the chart) propelled U.K. politics into the limelight once again.

Expectations of an increased majority for the incumbent Conservative Party, which was seen as making a “soft” Brexit more likely, initially helped sterling appreciate against the euro. The impact of the currency movement was evident from the strong performance of the S&P United Kingdom Index (in pounds sterling) in May compared to the same index denominated in euros. The subsequent tightening of the polls (point C) and the realization that no political party had secured a Parliamentary majority saw a complete reversal in this trend.

The posts on this blog are opinions, not advice. Please read our Disclaimers.

Inflation and the Fed

Contributor Image
David Blitzer

Former Managing Director and Chairman of the Index Committee

S&P Dow Jones Indices

Despite today’s very low inflation the Fed keeps raising interest rates and is now discussing when to shrink its balance sheet to further tighten monetary policy.  The Fed’s own inflation forecast anticipates continued low inflation at 1.6% in 2017 creeping up to 2% in 2018 and 2019 and not seeing any increase later on. Moreover, consumers and the public have similar expectations for low inflation now and in the future. So why is the central bank pushing interest rates higher?

The principal reason is the belief that a low and falling unemployment rate is a reliable signal of future inflation.  This is part of the received wisdom of economic policy which has some basis in the data. It originated with work done by a New Zealand economist, A. W. Phillips, in 1958 comparing unemployment and changes in wages in Great Britain from 1861 to 1957.  Phillips found a consistent relation that showed lower unemployment rates were associated with higher rates of wage inflation.  While the original finding was empirical – based on the data rather than economic theory – the idea soon became the basis for various theories of what causes inflation and how economic policy could lower either inflation or unemployment.

Current data on the U.S. economy since 2000 reveals similar patterns. The first chart shows the unemployment rate and the inflation rate (measured as the year-over-year change in the Core Personal Consumption Expenditures deflator) from January 2000 to May 2017.  As shown, when unemployment rose in 2008, inflation fell, suggesting an inverse relation.  The second figure is the Phillips curve of inflation and unemployment. It plots the unemployment rate on the horizontal axis and the inflation rate on the vertical axis. The regression line indicates that a one percentage point increase in unemployment is associated with a one-tenth of one percentage point decline in inflation.

The Phillips curve doesn’t prove that falling unemployment causes higher inflation. However the idea of a trade-off between inflation and unemployment is embedded in a lot of economic thought. Economists who remember the high inflation of the 1970s and the pain of the 1980 and 1981-2 recessions when unemployment surged and inflation collapsed pay attention to the Phillips curve. The experience of the last 12 months when both inflation and unemployment fell together isn’t enough to erase the Phillips curve.

Based on a combination of experience, data and economic theory, the Fed believes that continued declines in unemployment point to higher inflation rates. At some point the unemployment rate will hit bottom, inflation will be driven higher and a series of rapid increases in interest rates will be necessary. The result of that move could be a recession.

A related argument for gradual increases in interest rates is to prepare for the next recession: Were the economy to contract and the unemployment rate to rise, the Fed would want to lower interest rates. With the fed funds rate at 1.0%-1.25% and ten year treasuries trading at 2.4%, there isn’t much room to cut.  If the Fed slowly pushes interest rates back to more normal levels with the Fed Funds rate at 3%, its ability to stimulate the economy during a recession would be improved.

The posts on this blog are opinions, not advice. Please read our Disclaimers.

Asian Fixed Income: The Birth of Bond Connect

Contributor Image
Michele Leung

Former Director, Fixed Income Indices

S&P Dow Jones Indices

As a follow up to the previous article, Bond Connect officially launched on July 3, 2017. Bond Connect allows international market participants to trade China’s interbank bonds through the Hong Kong Stock Exchange.  It marked a milestone in China to further open up its capital market, following the China Interbank Bond Market (CIBM) announcement last year. Let’s take a closer look at the recent developments of China onshore bond market.

Similar to other existing channels, like RQFII and CIBM, Bond Connect allows market participants to access China’s onshore bond market, yet through the trade custody and settlement infrastructure connect features, Bond Connect is more cost and time effective for some market participants. As expressed by Bond Connect Company Ltd*, it significantly reduces the account opening time to three working days and improves the price discovery with electronic trading.

According to Mr. Pan*, Deputy Governor from People’s Bank of China, a total of 479 financial institutions invested in the China bond market with over RMB 800 billion under several schemes. The participation of overseas market participants is still low, around 3.9% in government bonds and 1.2% in overall bond markets.  It is unparalleled to the size of the China bond market, which was RMB 53 trillion as tracked by the S&P China Bond Index, hence there is further room for internationalization.

In terms of market performance, the total return of the S&P China Bond Index fell -0.49% year-to-date (YTD), while its yield-to-maturity tightened 23 bps to 4.31% during the same period, data as of July 3, 2017.  The index currently tracks the performance of 9,650 government and corporate bonds from China.  The S&P China Government Bond Index represents over 66% of the overall exposure, with a market value of RMB 35 trillion.

The S&P China Corporate Bond Index has expanded rapidly in the past 10 years, as the market value tracked by the index was RMB 18 trillion, which has increased 34-fold since the index’s first value date on Dec. 29, 2006, and the yield-to-maturity stood at 5.04% with a modified duration of 2.44 (see Exhibit 2 for the yield comparison).

On the back of the growth story, attractive yields, and diversification, the allocations in China’s onshore bonds are poised to rise in the coming years, particularly with the new accesses through Bond Connect and CIBM.

Exhibit 1: Market Value of the S&P China Corporate Bond Index and S&P China Government Bond Index

Exhibit 2: Yield-to-Maturity of the S&P China Corporate Bond Index and S&P China Government Bond Index

*Source: Bond Connect Investor Forum on July 3, 2017.

The posts on this blog are opinions, not advice. Please read our Disclaimers.