When inflation is taken into account, SPX, Dow and Nasdaq Composite indexes are still below their 2000 peaks. For some reason, no one talks about this.
this market is going to be on a tear for a while
Using the 2000 peak is not appropriate as we were clearly in bubble territory at that time. The market kept ripping up for 4 years after Greenspan’s irrational exuberance statement.
What inflation?
If you add in dividends, I’m almost certain sure it is higher, even after taking inflation into account. What is the rationale for including inflation but not dividends in your adjustment?
With nominal indexes there is at least the argument that these are the numbers that people see everyday and that people psychologically anchor on them. But if you are going to be “sophisticated” and discount for inflation, why not also be sophisticated enough to include dividends…???
I don’t think he made the chart, Bchad.
I’d argue dividends should be excluded from the analysis. The index is a measure of the value of its consistuents (arguably). The value of a firm, or index, does not include dividends paid in the past. He isn’t looking at returns here, he’s looking at equity value in the index, adjusted for CPI (which is a terrible measure to use for this purpose, IMO).
I realize that. I’ll rephrase the question as: “Why pay attention to a chart that adjusts for inflation but not dividends?”
That’s an intersting take. I generally look at a chart like this and think in terms of returns, which is why the dividends issue popped up so quickly with me.
At some level, one can ask why should it matter whether business assets are worth more or less in aggregate than they were in 2000 or 2008. They’re really all just psychological watermarks, since it’s forward looking returns that actually matter to what we do now. I think real-total return is the one that matters most because that’s how an investor who follows a passive strategy ultimately fares, though ultimately it would need to be tax adjusted for individual investors.
Wouldn’t inflation adjusted asset prices be expected to be mean reverting, excluding changes in the productive capacity of those assets? I think that’s one of the main things GMO uses to spot bubbles – the inflation adjusted prices of all sorts of asset classes.
Geo, What would you use instead of CPI?
Asset price growth rates may mean revert (in real terms or otherwise), but not asset price levels, which is what the chart tracks.
Mean reverting levels would imply that there is no real growth in company assets, and that any equity growth on balance sheets or market capitalization is purely an artifact of inflation.
My understanding is that GMO uses a standard of 2 standard deviations above the long-term trend as its operational definition of a bubble.
RE: dividends. Depends, on what you are looking for. If you are interested in pure investment return, i.e. “would I have made money if I bought equity in 2000”, then you would want to include dividends. If you are looking at stocks as an indicator of future earnings or even future economic output, then you should not include dividends.
Edit: Looks like someone already said something along these lines.
So, assuming that you are looking at stocks as an indicator of future earnings, what is the significance of stock indexes being above or below their 2000 levels in real terms, and why should we care about this more than the nominal number?
Some kind of asset price index… which I know is circular when looking at the prices of stocks… but CPI doesn’t really reflect the value of all assets, just a segment of assets (consumable ones).
Look at real-estate or other asset classes… they don’t really appreciate in line with CPI.
Tough to say what would work here, honestly. I don’t think CPI is an awful measure here, it’s just far from perfect. I don’t have a better suggestion in terms of anything that exists today as an index.
The idea that future expectations of real earnings is approaching levels of previous bubbles should be concerning, no? Do we think the world is heading ot even better growth than we did at the very peak of 2000? I certainly hope not. But that’s what this graph is showing, somewhat.
If this were something like a P/E chart or a P/B or Tobin Q, I’d agree with you, but there is nothing in this chart that involves any kind of forward estimate of profits/earnings/cash flows/etc. and therefore there is no way (from this chart alone) to determine whether things are expensive or cheap.
If it can’t be used for forward projection, then its real usefulness is for returns history, but that should include dividends.
It also may have psychological significance, but for that, the nominal price chart is almost certainly more important.
My point is just that the “stock market reaching new highs” headlines should have some qualifications. You can take as little or as much away from this as suits you.
Geo: “adjusted for CPI (which is a terrible measure to use for this purpose, IMO).”
Geo: “I don’t think CPI is an awful measure here, it’s just far from perfect. I don’t have a better suggestion in terms of anything that exists today as an index.”
Come on man, don’t roll over like that. The proper inflation measure depends on the indended use of the data. CPI should be used for adjusting payments to consumers of some basket of goods or services in today’s prices. IMHO, using a purely price return for these indices (which, incidentally, are reconstituted and manipulated every year) and then adjusting for CPI makes absolutely no sense in for any use of the data. This chart is basically saying that a certain basket of securities in 2000 is worth more than a different basket of securities in 2013 when you adjust for the price of cars, pants, cancer treatments and other sundry items.
The only way this chart makes sense is as a measure of returns to consumers who have to purchase stuff reflected in CPI, and that requires including the effect of dividends. Really, there should be two charts: one with dividends reinvested and one with the dividend yield disbursed.
Edit: agree with ohai that the “market reaching new highs!!!” gets too much hype, but that’s financial entertainment news. I am certain the person making this chart wanted to make this point and chose price return instead of total return for the index. Data manipulation 101.
If you don’t use the CPI, the alternatives tend to be to use the GDP deflator, the PPI, or to create some index of your own, typically linked to something like a commodities basket and possibly some foreign currencies. Justifying what you put into that that basket and what weights to use is usually a lot of work.
I usually go with the CPI, but I could see an argument for PPI or GDP deflator here. I’m not sure there is a huge difference.
Presumably. which index you use depends on who you are doing the analysis for. A typical family is affected by consumer prices a lot, so CPI makes sense. A business is faced with wholesale prices, so PPI might make sense. GDP is just a general alternative index for CPI. I used to remember better how it was constructed and I am too lazy to google it right now.