Stocks in Like a Lion, Out Like a Lamb?

Newfound Research  |

What was the old saying about March? “In like a lion...” You know it was a rough start when stocks were the asset of safety. While global equities were down only 1.84%, 20+ US Treasuries were down 4.47%, broad commodities were down 3.30% and US REITs were down 3.60%.

One bad week and we’re already seeing articles popping up saying “correction or bear Market?” However, the reality is that these micro-correlation storms happen from time to time, when suddenly the winds of global risk factors howl and every asset class moves in lock step.

Below we’ve plotted some weeks in the past when this same phenomenon has happened.


You’ll notice that micro-correlation storms tend to actually be positive events rather than negative ones. While only 1.47% of the weeks in this same period were weeks when all assets lost value, 4.44% of them were weeks when they all gained value.

What did those previous negative weeks look like at an asset-class level? We break them down below.


There are a few things we want to point out.

Firstly, this week’s loss level was not the worst of the bunch. It was bad, but not as bad as June 23, 2013. We can sum up all the asset returns to get an estimate of the aggregate magnitude of losses.

Even November 14, 2010 was a bigger global asset shock.

Secondly, these events are pretty random. Can you remember what actually happened the weeks of 6/23/2013 or 11/14/2010 that caused these losses?

Using Google, we can rewind the clock to find that the point-in-time narrative for June 2013 was a mix between taper concerns and a credit crisis in China. Of course, taper concerns had actually manifested earlier in the month based on comments from Ben Bernanke on May 22nd, so this week may have been nothing but market capitulation.

At least the weeks leading up to 6/23/2013 were fairly negative. November 14, 2010, on the other hand, had all the makings of a positive week, with strong labor market indicators and upside growth surprises in Europe. In fact, the week prior, the S&P 500 was up over 3.50%. The broad losses were largely blamed on economic growth concerns in China and threats that China may raise interest rates, which caused commodity markets to tumble. Yet just a few weeks later, the S&P 500 was up 3.44% and markets had moved on.

Finally, we’d be remiss if we didn’t point out that not a single one of these weeks occurred during the credit crisis. For the tremendous convergence in global asset correlations we saw during that period, not once did all the assets used in this study simultaneously exhibit losses in the same week.

Micro-correlation storms are shocks – and fierce ones at that. At the end of these sorts of weeks, we may look at our well-diversified portfolios and wonder if it is all for naught. History has taught us, however, while these squalls are full of sound and fury, they ultimately signify nothing.

In Our Models

There were no changes in our models this week.

We continue to see mixed signals. Under the hood of our Risk Managed Global Sector strategy, 5 sectors are currently exhibiting positive momentum, 3 are exhibiting neutral momentum, and 3 are exhibiting negative momentum. While the disposition of the signals may seem moderately bearish, digging in even deeper, we can see that those negative and neutral signals are much closer, on average, to turning back on than the positive signals are to turning off.  We estimate that it would take another 5-7% sell-off in global equities before our portfolio began to build a short-term U.S. Treasury position.

Sector strength is more positive in U.S. equities. 7 of the 9 sectors currently have positive momentum signals. Nevertheless, similarly to global equities, we estimate that only a 5-7% sell-off would be required for enough sectors to turn negative for the portfolio to begin to build a short-term Treasury position.

One interesting evolution that has occurred within these models is that the dynamic window that governs the amount of data used to estimate the momentum in each sector is, on average, about 22% shorter for the global sectors than the U.S. sectors. Therefore, while current estimates show that both models would begin to build a similarly sized position in short-term Treasuries with a similarly sized sell-off, we estimate the Risk Managed Global Sectors portfolio would continue to build a more significant position in a shorter time-frame if the sell-off then continued. 

ETF 02/27/2014
to 03/06/2015
to 02/27/2015
to 02/27/2015
1-3 Year U.S. Treasuries SHY -0.15% -0.29% 0.33% 0.18%
7-10 Year U.S. Treasuries IEF -1.84% -2.47% 1.72% -0.15%
20+ Year U.S. Treasuries TLT -4.47% -6.14% 3.07% -1.53%
Barclay's U.S. Aggregate AGG -1.18% -0.89% 1.14% -0.05%
Corporate Bonds LQD -1.91% -1.42% 2.28% 0.32%
High Yield Bonds HYG -0.94% 2.23% 2.95% 1.98%
S&P 500 SPY -1.50% 5.62% 2.49% 0.95%
Russell 2000 IWM -1.15% 5.95% 2.47% 1.30%
MSCI ACWI ACWI -1.84% 5.51% 4.12% 2.21%
MSCI EAFE EFA -1.98% 6.34% 7.00% 4.88%
MSCI EEM EEM -3.63% 4.41% 3.69% -0.08%
Commodities DBC -3.30% 4.43% -1.52% -4.77%
Gold GLD -3.70% -5.91% 2.27% -1.51%
U.S. REITs VNQ -3.60% -3.67% 2.93% -0.78%
International REITs VNQI -2.84% 3.68% 6.43% 3.40%
Global Conservative AOK -1.21%% 0.92% 1.29% 0.06%
Global Moderate AOM -1.26% 1.59% 1.56% 0.28%
Global Aggressive AOA -1.75% 4.18% 3.46% 1.64%


DISCLOSURE: The views and opinions expressed in this article are those of the authors, and do not represent the views of Readers should not consider statements made by the author as formal recommendations and should consult their financial advisor before making any investment decisions. To read our full disclosure, please go to:



Symbol Last Price Change % Change










Level Headed at World Crypto Con - Steve Beauregard CRO Steve Beauregard joins Jack Brewer at World Crypto Con.

Emerging Growth

Radient Technologies Inc.

Radient Technologies Inc is engaged in research, development and commercialization of environmentally responsible technology for the extraction, isolation and purification of soluble products from a wide range of materials using…