Merrill Lynch quantitative strategist Savita Subramanian is predicting continued market volatility using a chart I’ve called “the most interesting in finance.” The apparent ability of the yield curve to predict market volatility is a potential magic bullet for investors that signals they should start de-risking their portfolios immediately to protect themselves from a downturn.
The accompanying chart compares the steepness of the U.S. yield curve – which measures the gap between the yield on the 2-year and 10-year bond yields – with the CBOE Volatility Index (VIX Index; which I refuse to call by the hackneyed nickname 'Fear Index’), lagged by 36 months.
FORECASTING POWER
Can the yield curve predict market volatility?
U.S. yield curve steepness: 10Y bond yield
minus 2Y bond yield, in basis points
(left scale, inverted)
CBOE Volatility Index (VIX): lagged 36 months
(right scale)
-1 0 0
70
-5 0
60
0
50
50
40
10 0
30
15 0
20
20 0
10
25 0
30 0
0
1990
‘94
‘98
‘02
‘06
‘10
‘14
‘18
Oct.
Feb.
April
June
Aug.
Oct.
Feb.
Sept.
THE G L OBE A ND MAI L,
SOURCES: SCOTT BARLOW; BLOOMBERG
FORECASTING POWER
Can the yield curve predict market volatility?
U.S. yield curve steepness: 10Y bond yield minus 2Y
bond yield, in basis points (left scale, inverted)
CBOE Volatility Index (VIX): lagged 36 months
(right scale)
-1 0 0
70
-5 0
60
0
50
50
40
10 0
30
15 0
20
20 0
10
25 0
30 0
0
1990
1994
1998
2002
2006
2010
2014
2018
Oct.
Feb.
April
June
Aug.
Oct.
Feb.
Sept.
THE G L OB E A N D MAIL, SOURCES: SCOTT BARLOW; BLOOMBERG
FORECASTING POWER
Can the yield curve predict market volatility?
U.S. yield curve steepness: 10Y bond yield
minus 2Y bond yield, in basis points
(left scale, inverted)
CBOE Volatility Index (VIX):
lagged 36 months
(right scale)
-1 0 0
70
-5 0
60
0
50
50
40
10 0
30
15 0
20
20 0
10
25 0
30 0
0
1990
1994
1998
2002
2006
2010
2014
2018
Oct.
Feb.
April
June
Aug.
Oct.
Feb.
Sept.
THE G L OB E A N D MAIL, SOURCES: SCOTT BARLOW; BLOOMBERG
The lag means that the VIX level is plotted only as recently as Sept. 30, 2015. In the end, the chart is an attempt to gauge whether the yield curve has been able to predict the VIX index level three years later. Higher VIX levels are usually associated with market sell-offs, so if we trust the chart, investors would be able to predict market weakness well in advance and reduce risk in their portfolios before it happens.
There are problems with the statistical relationship. For every time period, such as June, 2007, to February, 2015, where it looks very much like the yield curve correctly forecast the VIX, there are others, such as June, 1996, to June, 1999, where the two lines on the chart have seemingly nothing to do with each other.
Still, credit expansion gives support to the idea that the yield curve and market returns are related.
Profits on lending – where, for example, a bank borrows at short term-interest rates and lends to clients at longer-term interest rates – are determined to a great extent by the yield curve. The steeper the curve, the bigger the bank’s profits because they are determined by the difference between a lender’s cost of funds (short-term rates), and the interest payments they receive from borrowers (at longer-term rates).
The connection between the yield curve and the VIX then would go like this: A flattening yield curve means lender profits are declining, so they reduce loan activity and eventually, in 36 months, this starves corporations of funds, leading to falling equity markets and a rising VIX.
Remember that the purple line on the chart (VIX) is lagged three years, so where it goes from there is where the VIX goes from now. If the relationship between the yield curve and the VIX is really predictive, the VIX is set to move significantly higher in the next three years (the purple line will follow the blue line). This implies weak U.S. equity market returns for the foreseeable future, and that investors should start reducing risk in their portfolios.
I honestly don’t know whether to trust the chart or not. It is tempting to believe it because it would be an extraordinarily powerful investment tool, allowing for the accurate prediction of bear markets – at least approximately.
Ms. Subramanian, importantly, does believe in the forecasting power of the yield curve. The strategist has experience and access to resources well beyond mine and is someone I confidently put on the list of top five U.S. equity strategists.
The chart, combined with other signs of market risk such as the slowing global economy, extended age of the rally, and rising interest rates, is enough of a catalyst for investors to partly reduce risk in their portfolio. This might involve taking some profits on high-valuation equity holdings, or otherwise increase allocations to cash.
Scott Barlow, Globe Investor’s in-house market strategist, writes exclusively for our subscribers at Inside the Market.
Scott Barlow