The 'good-is-bad and good-is-good', or as Morgan Stanley's Adam Parker calls it, the "Hall-Pass' market is one of four regimes that investors face in the current environment. In the current world, he finds that negative economic surprises - while causing negative short-term responses in stocks - are rapidly mean-reverted into a positive return reflecting this 'good/good' response suggesting participants now viewed tapering as the base case: good data would presumably help the post-taper economy, while bad data might lead to a delay or mitigation of the tapering. The market has been mostly in this 'Hall-Pass' mode since the start of 2013 but fell briefly into 'normal' mode when Taper talk began in May (until Bernanke and his cohorts jawboned us back from the edge). Critically though, Parker notes, while the current period is also one where the market responds favorably to both directions of economic surprises, the drift in responses is now flat to down: in the absence of large economic surprises, we would therefore expect the market to be flat to down.
Via Morgan Stanley's Adam Parker,
Our main interest is in the sign of the response (i.e., estimated coefficient) to positive and negative surprises. Based on the four combinations of signs, we obtain four “regimes” as described in the chart below:
While the US equity market is liquid, it may still under- or overreact to economic surprises. We therefore consider the impact of one-day lagged positive and negative changes in the economic surprise index on the equity market. As shown in the chart below, one-day lagged positive economic surprises are only (10%) significant at a rate consistent with luck (i.e., matching the 5% unconditional probability of both tails). Meanwhile, negative economic surprises show signs of mean reversion in the day following their release: On the day of the release, the coincident negative coefficient is positive and frequently significant, indicating that the market tends to sell off on the news. The one-day lagged negative coefficient is frequently negatively significant, however, indicating a (partial) bounce-back from yesterday’s data. This suggests over-reaction by the equity market to bad economic news, followed by moderation.
In the chart below, we show the resulting regimes over time. Normal markets are shown with a value of +1; periods where markets respond favorably to good and bad economic news are denoted with a +2; periods when the market responds unfavorably to both types of news are denoted with a -2; and periods where the response is opposite of normal are indicated with a +3. While this latter state is not “better” than the good/good state, we wanted these abnormal periods to stand out on the chart.
Over the last 10 years, about 1/3 of 63-day periods are in the normal state, where equities rise on positive surprises and fall on negative surprises.
In 2013, the equity market has primarily had either a good/good or normal response to economic surprises. Once QE3 and QE4 were established in December 2012, and the Fiscal Cliff had been dealt with, markets responded favorably to both positive and negative surprises until early April 2013.
We have noted that QE began to lose its efficacy around this time, and the April 3 transition date to a normal response is consistent with this observation. The normal response persisted through the Fed change in communication regarding possible tapering of QE, but recently (July 5) the market returned to a good/good state. The current good/good response would be appropriate if participants now viewed tapering as the base case: good data would presumably help the post-taper economy, while bad data might lead to a delay or mitigation of the tapering.
Net of effects due to economic surprises (or in the absence of economic data), the equity market has generally tended to drift upwards, with 67% of periods having positive intercepts since 2003. During 2013, the market drifted upward from late January through most of June, before turning negative in recent days. Through February and March, the market had tail winds of positive drift and good responses to good and bad economic data.
While the current period is also one where the market responds favorably to both directions of economic surprises, the drift is now flat to down: in the absence of large economic surprises, we would therefore expect the market to be flat to down.