財經觀察 2163 --- 50D MAVG vs 200D MAVG

I walked into the office yesterday evening and there was someone on CNBC talking about how the 50-day moving average of the S&P 500 rising above the 200-day moving average was telling us the market was getting ready to rise and the recovery had started. I listened to his babbling for another 2-3 minutes and couldn't take it anymore (and no, it was not my friend Larry Kudlow, who is a lot more balanced than whoever was on.)

We keep getting told that the market is telling us "something," usually that the recession is going to end. For some reason, people keep repeating the bromide that the market looks out about 6 months. To that I politely say, rubbish.

Riddle me this, Batman. Did the market see the recession in October of 2007? We were already in recession and the S&P 500 (see below) was making new highs! Where was the market prescience? Did it see the 25%+ drop in January of this year? And I could go back and cite scores of examples where the market "missed" the future turning points over the past ten decades.

What about the shibboleth that the market turns up 6 months before the end of a recession? Sometimes that is true. But does it mean anything? The same people who said it meant something last December and January are saying it means something now. But now it's June and the recovery is not here, so maybe the market wasn't telling us something in January after all.

Gentle reader, there will be a recovery. We will talk about what kind in a few pages, if we have the time. And it is (statistically speaking) likely that the markets will have turned up before the actual recovery. But does that mean anything today?

Go back to the chart above. Notice that in 2003, when the market finally turned up, we were already well out of recession. And the market had a very quick 12% or so drop while we were in recovery, while later we went on to a 90% run-up! Was the drop telling us anything, or do we explain it away?

"In the short run," St. Graham said, "the market is a voting machine. In the long run it is a weighing machine." The voting is based on current sentiment, but what the market weighs in the long run is earnings. The market tries to forecast future income streams. And it gets it wrong as often as it gets it right.

Let's look at this yet another way. This is an important concept, and it should be a component of your economic BS detector. The CNBC host talked in breathless terms about the importance of the 50-day average moving above the 200-day average. It means nothing until it means something, and we won't know what that something is for some time.

Earlier this week (Monday, I think) the 50-day average moved BELOW the 200-day average. The analysts at Bespoke Investment Group noted:

"Going back to 1928, this is the 25th time that the S&P 500 has declined through both of these levels on the same day. On page two we have provided a table showing each of these occurrences as well as the index's returns going forward. Based on those prior instances, the S&P 500's returns going forward have been notably negative. While the S&P 500 has averaged positive returns over the next week, average returns have been negative over the next month, three months, and six months." (emphasis mine)

But 33% of the time, the markets were up six months later, often by quite a bit. And sometimes down quite a bit, but on average only slightly. Which means that as a forward-looking indicator it is interesting but not anything I would put my money (or client money) on!

(I saw some reports that differed, selecting fewer such data points and suggesting that market returns were up after such an event. Logically, that can't be. Let's be generous and just assume sloppy research.)

Before major market moves down, the 50-day average will always move below the 200 average. And the reverse is also true. It is not a sign. It is just what statistically MUST happen. And sometimes they reverse themselves, and sometimes they don't. We have no way on God's green earth of knowing whether the two moves (both up and down) this week will be bullish or bearish six months from now, based simply on the moving averages crossing. You can make the data say anything you want, but you are still just guessing.

Sidebar note: Trend Following 101. I spend a lot of time analyzing trend-following money managers of one kind or another. Basically, they look at data and try to spot trends and then invest in them. A trader who is right 70% of the time is amazing and very rare. 50% is more like it for successful traders. But they have sharp risk controls that cut their losing trades and let their winning trades "ride." Being right 50% of the time can be profitable over time. (Being right 50% of the time is harder than it looks!)

But in the media you get these "analysts" who talk a good game, acting as if a 50-70% probability is something meaningful. "The market has turned. The recession is over." And they say that when we have the first balance-sheet recession in 70 years, yet they want to compare garden-variety recessions to what we have now. Again, we can only know which of the moves (above and below the 200-day moving average) will be the real "indicator" in six months. It is only an indicator today to the extent that we can drive our cars forward looking in the rear-view mirror.