Morningstar and the Scam of Forward-Looking Earnings

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Topic Author
berntson
Posts: 1366
Joined: Mon Oct 29, 2012 12:10 pm

Morningstar and the Scam of Forward-Looking Earnings

Post by berntson »

Morningstar uses forward-looking earnings to calculate things like price-to-earnings, price-to-book, and so on. That is, they are using some sort of financial dark-magic to forecast the earnings for the coming coming year, then using that to determine stock valuations. These valuations are then used to generate the famous Morningstar style boxes.

If Morningstar is going to use forward-looking earnings, one would at least hope that those estimates were generated in some simple and fairly transparent way. Maybe they take last year's earnings and multiply it by the average growth rate over the last five years. But as far as I can tell, their forward-looking earnings are generated by some sort of opaque, proprietary process that throws the opinions of analysts into a blender and then spits out a number.

Morningstar then presents the results of its entrail readings as if they were accurate to a high degree of precision. It claims that Vanguard total market has a P/E of 15.85 for example. There are no error bars and no indication that this is at best a stab in the dark.

A further problem with Morningstar's use of forward-looking earnings is that it biases measurements like P/E downwards. Since companies will typically have more capital in the coming year than in the last, they will typically have higher earnings in the coming year than in the last (in the same way that your savings account will typically have higher earnings in the coming year than in the last because of compound interest). So forward looking P/E is lower than backwards looking P/E. This isn't a problem for careful investors who read footnotes and are careful about such things. It is a problem, though, for the naive investor who are not always careful to keep track of whether they are using forward or backward valuations.
Buysider
Posts: 764
Joined: Sun May 18, 2008 9:36 pm

Re: Morningstar and the Scam of Forward-Looking Earnings

Post by Buysider »

Forward earnings are the only thing that matters. The past is, at best, only a poor predictor of future earnings. Investors care about future earnings, cash flows and dividends from their investments, not what they paid in the past.
There are no error bars and no indication that this is at best a stab in the dark.
Morningstar is just using the consensus estimates of analysts for forward earnings. These are extremely accurate ... for what they are. You can say they in fact are the market's expectations for future earnings AT THIS POINT IN TIME. There is no error bar (though you can look up the earnings estimate dispersion if you are so inclined) - the estimates are the estimate. As the course of the year goes on, the estimates may change, as the facts about the economy, company, competitors, etc. change.

Tesla made about $0.70 last year and is expected to earn $1.64 this earn in adjusted eps. Which number is more relevant for you in determining whether or not to buy the stock?
Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: Morningstar and the Scam of Forward-Looking Earnings

Post by Rodc »

These are extremely accurate
What does that mean? How many significant digits on average, for example?

Estimates, like measurements in general, with no notion of error bars or confidence intervals have very little usefulness.

Compare company A and company B.

One has a P/E of 15 and that other a P/E of 12, which should I buy?

If it is 15+/-5 and 12+/-6 I might want to invest in both as either might be a good or a bad buy and I want to spread my risk.

If it is 15+/-0.5 and 12+/-0.6 I have an entirely different story.

And if in addition I have a bias term, I better know that as well.

I agree completely that I want to know forward earnings. But just because want to know them does not mean I can.
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.
Topic Author
berntson
Posts: 1366
Joined: Mon Oct 29, 2012 12:10 pm

Re: Morningstar and the Scam of Forward-Looking Earnings

Post by berntson »

Buysider wrote:Forward earnings are the only thing that matters. The past is, at best, only a poor predictor of future earnings. Investors care about future earnings, cash flows and dividends from their investments, not what they paid in the past.
I agree that investors care about about future earnings.

The problem is that Morningstar presents its forward looking p/e, p/b, p/c, etc. ratios as "Value & Growth Measures". These aren't measures, these are analyst-driven estimates. Investors are lead to believe that they are getting some sort of quantitative,vaguely objective measurement of a company's value, when they are really getting an aggregate analyst estimate of a fund's future performance. Morningstar might as well the price to buy-rating ratio while they're at it.
User avatar
nedsaid
Posts: 19275
Joined: Fri Nov 23, 2012 11:33 am

Re: Morningstar and the Scam of Forward-Looking Earnings

Post by nedsaid »

Bernston, you raise a great point that a lot of investors don't think about. When you see a P/E ratio, you need to know if they are based on future estimated earnings or on the past year's earnings. Looking forward or looking backward.

I believe that the markets look forward and so I tend to look at P/E's based on estimated future earnings. But one should be aware that earnings estimates from Wall Street analysts are often wrong.

But I wouldn't say that forward-looking earnings are a scam. But they are an imperfect measure.
A fool and his money are good for business.
User avatar
G-Money
Posts: 2867
Joined: Sun Dec 09, 2007 6:12 am

Re: Morningstar and the Scam of Forward-Looking Earnings

Post by G-Money »

Same could be said for inflation. I don't believe, for example, that either the Cleveland Fed or Philly Fed publish error bars for their estimates of future inflation. Perhaps not.

I don't put much faith in predictions anyway, and I'm not as math savvy as many of the folks here, so it doesn't bother me much if error bars etc. aren't included.
Don't assume I know what I'm talking about.
Buysider
Posts: 764
Joined: Sun May 18, 2008 9:36 pm

Re: Morningstar and the Scam of Forward-Looking Earnings

Post by Buysider »

Bernston wrote:
These aren't measures, these are analyst-driven estimates.
Analyst estimates of future earnings are the best source of data we have on what the market's expectations of earnings are over the next 1-2-3 years are for companies. In Japan, companies publish their own forecasts, but, not surprisingly, analysts estimates are more accurate.

Investing isn't like physics, and next year's earnings aren't even that relevant for the value of a stock (it is mostly driven by the cashflows over the next 10-15 years). But, we have to use the data we have, and data that incorporates more information is generally better than data that doesn't include more information.

To give another example, Tepco (owner of Fukushima) - trailing earnings were almost completely irrelevant post the 2011 accident.


Rodc wrote:
Estimates, like measurements in general, with no notion of error bars or confidence intervals have very little usefulness.
The high tomorrow in my town is forecast to be 40 degrees. I know I'm not going swimming and I have no idea what the std deviation of the forecasts are. Earnings estimates, likewise, don't convey much information - they just convey more than historical reported earnings. A PE of 15 means nothing alone, and doesn't mean a stock is better or even cheaper than a stock that has a PE of 20.
MN Finance
Posts: 1926
Joined: Sat Dec 22, 2012 9:46 am

Re: Morningstar and the Scam of Forward-Looking Earnings

Post by MN Finance »

What a scam. They should really get in line with the rest of the industry so there's some consistency.... oh, wait.
gisborne
Posts: 15
Joined: Wed Jan 16, 2013 8:14 pm

Re: Morningstar and the Scam of Forward-Looking Earnings

Post by gisborne »

Is this some kind of modern trend to make it look like P/E ratios are lower than they are relative to the past? When did this "forward looking earnings" thing start? I'm pretty certain if you look in most investment books that explain PE they don't say "divide the current price by your best guess of future earnings".
User avatar
nisiprius
Advisory Board
Posts: 52216
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Morningstar and the Scam of Forward-Looking Earnings

Post by nisiprius »

So you're saying Morningstar's style boxes are subjective? Whoa. I had no idea. If they are going to publish what are in effect predicted styles, then at the end of the year the least they should do is publish something that shows what the actual style category turned out to be, and how often the prediction was wrong.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: Morningstar and the Scam of Forward-Looking Earnings

Post by Rodc »

The high tomorrow in my town is forecast to be 40 degrees. I know I'm not going swimming and I have no idea what the std deviation of the forecasts are. Earnings estimates, likewise, don't convey much information
Fair enough. If you have an estimate of something that has almost no bearing on any decision or action you do not need to know how accurate it is.
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.
Scooter57
Posts: 2019
Joined: Thu Jan 24, 2013 8:20 am

Re: Morningstar and the Scam of Forward-Looking Earnings

Post by Scooter57 »

Many US companies do issue estimates of their next year and next quarter earnings. Some issue updates when conditions change. Analyst estimates use those numbers and others they seem to pull out of thin air. Typically they start out too optimistic and drop as the earnings report date approaches. However, I have more confidence in individual company estimates than I do in averages applied to a cap weighted index. That is just too much smoothing applied to vague numbers. And I always adjust individual company estimates downwards, too.

With an individual business you can ask, what is this estimate based on? Does it hold up to common sense scrutiny? Has this company a good history of meeting estimates? But with every company mooshed together and averaged to produce a number you lose that.

People love broad based measures--stuff like GDP--but my experience with those kinds of derivative numbers in my field of work has made me very suspicious and cautious about any conclusion based on averages, and even more so averages that have come out of statistical software that applies complex manipulations to the base data.
Post Reply