“Money flooded into stocks last week.” This inane headline greeted me this morning from my inbox. With certain people tipping into euphoria, and even the US president claiming credit for the stock bubble, it pays to keep reminding ourselves that the situation isn’t normal.
It’s pretty obvious that money can’t “flood” into any second hand investment. Any money inflows must be matched with outflows. For every buyer there is a seller.
Hence, headlines about money flooding into stocks make no sense. Certainly, one group of investors can be throwing money at the market. But that must mean that another group is catching the cash and moving to the sidelines.
Most likely, with stock valuations in bubble territory, there are some late entrants climbing onto the bandwagon. The philharmonic orchestra – the smart money – has played itself out and is heading home. It’s giving up its seats to the junior school band – the dumb money – which is keen as mustard, but lacking in technique. Screech, bang, honk, crash.
Why do I think US stocks are in a bubble? The evidence is in plain sight, for anyone that cares to remove their blinkers. Here’s a selection of what I’ve written about it this year:
- Is the US stock bubble as big as year 2000?
- US stocks back in 1929-style bubble
- Another (worrying) record for US stocks
Of course, just because a market is in a bubble doesn’t mean it can’t keep grinding higher. And that’s what this market keeps on doing.
In other words, the point at which marginal sellers are coaxed into selling, and how much marginal buyers are prepared to pay to tease them out, keeps edging upwards. That’s all a market price is: the level where the keenest buyers transact with the keenest sellers at a moment in time.
(For more on the different actors within the market mob, and how prices are set, see “Horses, dances and market prices”.)
How long can this go on?
How long can this last? How much higher can it go?
Nobody knows. But if you’re in it, you’d better get ready to get out fast when the time comes.
How much could the US stock market eventually fall? No one knows that either. But we can hazard a guess at the potential magnitude. The picture is never crystal clear, but that doesn’t mean it isn’t worth studying.
My earlier articles give some clues as to where we find ourselves (see the links above). Today we can squint at the problem from another angle.
The following chart looks at the level of the S&P 500 since 1871, adjusted for inflation and shown on a logarithmic (“log”) scale. Log charts are useful in investment charts that cover a long time scale. That’s because a vertical move of the same percentage size always looks the same to the naked eye.
For example, going from 100 to 150 is an increase of 50%, and so is the move from 1,000 to 1,500. Both look the same on this chart.
Stock prices tend to rise above the rate of inflation. In the case of the S&P 500, the average, compound, real (above inflation) price gain was 2.3% a year since 1871. Along with dividend income, this is what makes stocks such great long term investments (and great shorter term investments, if bought well).
Clearly there are market cycles. Real profit growth can be faster or slower than average in a given year. But if profits (strictly, earnings per share, or EPS) grew smoothly, at a fixed rate above inflation, and if P/E remained constant, the jagged line in the above chart would be dead straight. It would start at the bottom left and trend upwards to the top right.
When stocks are well below such a trend line, then it’s almost certain that they’re cheap. And when they’re well above the line they’re likely to be expensive. Just eyeballing the chart above, you can see that stocks were very cheap in the early ‘80s. You can also see that they’re most likely expensive today.
To check this, I’ve put the monthly real price data for the S&P 500 into a chart, and added the actual trend line (as calculated by a spreadsheet). This is what it looks like.
Indeed you can see the actual market is well above the trend. Today’s S&P 500 level is 2,568. The trend line is at just 1,240.
This means the market is currently trading at more than twice its long term trend. Or, put another way, it would have to fall 52% to get back to trend. Yikes.
Here’s exactly the same data on another chart, but with a linear scale instead of a log one (which makes the scale on the right easier to read). With a linear scale, the straight trend line of market growth becomes an exponential curve, and the movements in the index level look more pronounced as you moved from left to right (for the same percentage change).
Unfortunately, things get a little more complicated. I can’t just say the S&P 500 is twice where it should be, based on its historical trend (taking account of inflation).
Complicating factor: stock buybacks
A change in the law in the early 1980s made it easier for US companies to replace dividend payments with stock buybacks. Since buybacks reduce the number of shares, earnings are divided across fewer shares, and hence EPS increases.
The result is that present day EPS is higher than it would have been without the move to more buybacks. This in turn means the index is higher than it would have been.
Here’s a chart I put together a while back which shows the shift of cash payouts to shareholders from dividends (blue) to buybacks (red) over time.
This should mean that the trend line changes after about 1985, bringing it higher. But it’s not quite that simple. At the same time US companies have increased the amount of new stock issued to management and employees. This dilution of external shareholders offsets much of the supposed benefit of the stock buybacks.
For example, I previously calculated that in the ten years to 2015 the gross buyback yield was 3.2%. But new issuance was so high that the net buyback yield was only 0.2%. Similarly, for the five years to 2015, the gross buyback yield was 3%, but the net yield was just 0.5%.
Of course, there was always share issuance to raise new capital. So it’s difficult to know how much net effect to adjust for (the adjustment being gross buybacks less issuance for stock issued to insiders as pay). But let’s have a go.
Adjusting for a net effect of 0.75% a year since 1984 brings the adjusted trend line up to 1,587. At that level the index would have a cyclically adjusted P/E (CAPE) ratio of 19.4, which is very close to the median CAPE of 19.7 since 1970. (Since 1970 the market has seen full bull and bear cycles – between the extremes of 1982 and year 2000 – and the dollar was unpegged to gold almost all of the time.)
That’s still 38% below today’s actual market level. In other words, it’s still a big fall to trend, even with this big adjustment..
US stocks may keep grinding higher, or head sideways, or fall. It’s impossible to predict. But I challenge anyone to make a case for them being cheap at their current levels. I certainly haven’t seen one yet.
In the meantime, I urge caution for those invested in US stocks.
Stay tuned OfWealthers,