In an ideal world, portfolio investing is all about comparing the returns available among the three liquid asset classes–stocks, bonds and cash–and choosing the mix that best suits one’s needs and risk preferences.

In the real world, the markets are sometimes gripped instead by almost overwhelming waves of greed or fear that blot out rational thought about potential future returns.  Once in a while, these strong emotions presage (where did that word come from?) a significant change in market direction.  Most often, however, they’re more like white noise.

In the white noise case, which I think this is an instance of, my experience is that people can sustain a feeling of utter panic for only a short time.  Three weeks?  …a month?  The best way I’ve found to gauge how far along we are in the process of exhausting this emotion is to look at charts (that is, sinking pretty low).  What I want to see is previous levels where previously selloffs have ended, where significant new buying has emerged.

I typically use the S&P 500.  Because this selloff has, to my mind, been mostly about the NASDAQ, I’ve looked at that, too.  Two observations:  as I’m writing this late Tuesday morning both indices are right at the level where selling stopped in June;  both are about 5% above the February lows.

My conclusion:  if this is a “normal” correction, it may have a little further to go, but it’s mostly over.  Personally, I own a lot of what has suffered the most damage, so I’m not doing anything.  Otherwise, I’d be selling stocks that have held up relatively well and buying interesting names that have been sold off a lot.

What’s the argument for this being a downturn of the second sort–a marker of a substantial change in market direction?  As far as the stock market goes, there are two, as I see it:

–Wall Street loves to see accelerating earnings.  A yearly pattern of +10%, +12%, +15% is better than +15%, +30%, +15%.  That’s despite the fact that the earnings level in the second case will be much higher in year three than in the first.

Why is this?  I really don’t know.  Maybe it’s that in the first case I can dream that future years will be even better.  In the second case, it looks like the stock in question has run into a brick wall that will stop/limit earnings advance.

What’s in question here is how Wall Street will react to the fact that 2018 earnings are receiving a large one-time boost from the reduction in the Federal corporate tax rate.  So next year almost every stock’s pattern in will look like case #2.

A human being will presumably look at pre-tax earnings to remove the one-time distortion.  But will an algorithm?

–Washington is going deeply into debt to reduce taxes for wealthy individuals and corporations, thereby revving the economy up.  It also sounds like it wants the Fed to maintain an emergency room-low level of interest rates, which will intensify the effect.  At the same time, it is acting to raise the price of petroleum and industrial metals, as well as everything imported from China–which will slow the economy down (at least for ordinary people).  It’s possible that Washington figures that the two impulses will cancel each other out.  On the other hand, it’s at least as likely, in my view, that both impulses create inflation fears that trigger a substantial decline in the dollar.  The resulting inflation could get 1970s-style ugly.

My sense is that the algorithm worry is too simple to be what’s behind the market decline, the economic worry too complicated.  If this is the seasonal selling I believe it to be, time is a factor as well as stock market levels.  To get the books to close in an orderly way, accountants would like portfolio managers not to trade next week.