Is 95% of what you read in the financial press “either wrong or irrelevant?”

Question

From a Cato Institute article by Steve Hanke¹ (emphasis added):

Most press reports about Zimbabwe’s fantastic hyperinflation are off the mark – way off the mark. Even our most trusted news sources fail to get the facts right. This confirms the “95 Percent Rule”: 95 percent of what you read in the financial press is either wrong or irrelevant.

Is the 95 percent rule accurate? In other words, is "95 percent of what you read in the financial press is either wrong or irrelevant?"

Note: Steve Hanke certainly seems to believe that inflation for some countries is higher than actually reported (e.g. the Zimbabwe article I cited, this tweet regarding Sudan from 9 hours ago, and this tweet regarding Argentina from 11 hours ago). This may or may not influence his opinions about the financial press. This question has nothing to do with whether inflation (or hyperinflation) is incorrectly reported.

¹Professor of Applied Economics and Co-Director of the Institute for Applied Economics, Global Health, and the Study of Business Enterprise at The Johns Hopkins University in Baltimore

But this belongs on Skeptics! No, it doesn't.

@RupertMorrish Thanks for the link! Turns out this is a much more prevalent concept than I expected. I searched up "95 percent of what you read in the financial press is either wrong or irrelevant?" but didn't get relevant results. Guess I was searching to generally. — Barry Harrison, May 13 '19 at 06:57
@BarryHarrison So Sturgeon's law also applied to your search results? :) — nanoman, May 13 '19 at 07:09
That's true of any news. The currency of news is speed, not accuracy. — Harper - Reinstate Monica, May 13 '19 at 16:16
Who is "you" here? Perhaps 95% of what is printed is irrelevant, but I think I avoid a lot of that by a) not reading much of the financial press; and b) not actually reading an article unless it looks interesting. (Which of course is not the same as relevant.) — jamesqf, May 13 '19 at 16:38
"95% of all statistics are made up by the authors of the item you are reading." I leave it as an exercise for the reader to determine whether this item is one of the exceptions to the rule quoted above. — Dilip Sarwate, May 14 '19 at 02:52
I'm not sure if the opening paragraph you quoted has any real relevance to the claims the author is trying to make, but taken on it's own, it seems like nonsense. Specifically "This confirms the “95 Percent Rule”: 95 percent of what you read in the financial press is either wrong or irrelevant." I would love to hear how someone can use a single piece of data to confirm that 95% of all the data is innacurate. They must be really good at statistics if they can figure that out... — JMac, May 14 '19 at 16:40
@BarryHarrison This is the quote you wanted to question, so I'm not trying to recommend you quote anything different. Just pointing out that the quote is at best bad writing, since a single point of data would never be enough to "confirm" the statistics he references. — JMac, May 15 '19 at 10:51

score 38 · Accepted Answer · answered May 13 '19 at 06:09

38

If this is understood as "irrelevant to investment decisions", then it is plausible because publicly released news is immediately (in less than a second) incorporated into asset prices, according to the efficient market hypothesis. Thus (unless you are one of the big guns of Wall Street trading at light speed) by the time you read that a country or a company is doing well or poorly, it is too late for this to give you any insight on what you should buy or sell. The 5% that is relevant could be relatively timeless commentary on asset allocation, minimizing fees and taxes, etc.

answered May 13 '19 at 06:09

nanoman

29,638
3
69
89

3

For some types of news this is the case, for others, not at all. Most of the news that is relevant to investing, though, I would agree does not come from the financial press, but rather the press that is recording what is going on in the rest of the world where economics is actually unrolling day by day. Political news, industry news, the popular press, etc, all contain clues about what people are doing, what they're intending to be doing, what they'd rather be doing, and what they're upset about... these are all important financial clues. – J... May 13 '19 at 13:39
1

@J... The question (and thus the answer) is explicitly about "the financial press". – pipe May 13 '19 at 15:16
5

The "less than a second" claim seems unnecessary and implausible. There are plenty of market-shifting public events that take more than a second of human attention to recognise their market importance. Perhaps after whatever time it takes to form a particular judgement from the news has passed, this is incorporated into asset prices within a further second, but that's quite far from saying that news is only relevant for one second. Big news stories precipitate increased volumes and price volatility over the span of multiple trading days. – Will May 13 '19 at 17:01
@pipe Indeed, and that's why I made the statement explicit. It's an important caveat that, while financial news may be backward-looking (and while some minority of financial news may be forward looking) this doesn't necessarily mean that other news may not be useful for developing an investment strategy. – J... May 13 '19 at 17:02
5

While this may be true, it doesn't seem to be what he's trying say when he says "off the mark – way off the mark". He's not just saying that the information is too outdated to be useful, he's claiming that it's totally wrong. – Barmar May 13 '19 at 19:50
@Will Ain't that the truth! Witness any news event that takes place during After Hours trading, for instance, which causes the stock to spike, only for it to plummet at the open the next day, or vice versa. – May 13 '19 at 20:21
Forgot to accept. My bad! – Barry Harrison May 27 '19 at 05:38

JimmyJames · Answer 2 · 2019-05-14T15:39:19.877

11

While only the author of that statement can tell you exactly what was meant, I would relate it to a seemingly similar argument made by Nassim Nicholas Taleb in "Fooled by Randomness" that it's pointless to pay attention to the daily (or hourly) updates on what is happening in the news or markets. His assertion is that it's mostly noise and statements from the press like "the markets moved up 1% on jobs numbers" are purely conjecture because there is no way that the daily, (hourly, minute-to-minute) news cycle has empirical knowledge of why the markets moved one way or another.

On a side note, this specific article appears to prove it's own point. Essentially it says: "the Economist is wrong, look at this table created by me, see?" Even if you assume the author's numbers are better, look at how many zeros there are in the last two figures. The precision on this is really low and when you compound imprecise numbers they get much more imprecise. In other words, the article states that the author's analysis leading to an obviously imprecise figure is 'right' and that some other body's analysis is 'wrong'. But at this level of imprecision and lack of reliable information, the meaning of 'wrong' or 'right' is a fuzzy concept. And either of rates of inflation means that the currency is in a death-spiral. The difference (as large as it is) has no practical implications. You aren't going to decide to buy Zimbabwe dollars because you used the IMF numbers.

edited May 14 '19 at 15:39

answered May 13 '19 at 15:14

JimmyJames

3,840
11
19

4

"are purely conjecture because there is no empirical way to determine why the markets moved one way or another." Not only conjecture, but arguably meaningless. If there are ten things that move the market up 1 each%, and nine things that move it down 1% each, does it make sense to single out one of the things that moved the market up and say the rise is due to that? – Acccumulation May 13 '19 at 15:43
1

Are you seriously suggesting that there's no objective difference between an imprecise 12-figure number and an imprecise 23-figure number because both are imprecise? – sgf May 13 '19 at 15:46
2

And the Cato institute's complaint is that the Economist uncritically copies the IMF's faulty data, while they have published scientifically accurate (to a certain degree) data, which the scientific literature actually cites. They're not just saying "but we have other guestimates", which your answer seems to imply. – sgf May 13 '19 at 15:47
@sgf What I am saying is that I'm not seeing a lot of evidence to back up the claim that these are "accurate estimates of Zimbabwe’s fantastic hyperinflation". The linked article makes a case that they have a better methodology but it's still based on unofficial sources such as the "data for the black-market Zimbabwe dollar/U.S. dollar exchange". Accurate to how many 100's of quintillions (the implied precision) exactly? How is it that the the precision went from basis points to 100s of quintillions percentage in one month? And what practical difference does this make? – JimmyJames May 13 '19 at 17:25
"there is no empirical way to determine why the markets moved one way or another." That's not true. There might not be any way to prove why the markets moved one way or another (absent phoning up the people who sold stock and asking them,) but there are certainly empirical methods that can yield reasonable results. If the market moves 1% within 10 minutes following the biggest news item of the day, for example, the odds are pretty good that that was the primary reason. It's relatively rare to have more than 1 or perhaps 2 news items large enough to cause significant differences within a day. – reirab May 13 '19 at 22:30
@JimmyJames [Disclaimer: I did these numbers on an online calculator and didn't double check] Say their estimate is horrible and 80% off, i.e. the real number is .2 times of what they claim it is. That would make the Economist's number 3,588,000,000,000% off. You can't pretend that that is the sort of difference between guestimates where it's impossible to empirically verify that one of them is closer to the truth than the other. (Of course, that doesn't mean that the Cato institute's number is closer to the truth, I don't know how to actually arrive at those numbers.) – sgf May 14 '19 at 06:46
@sgf "that doesn't mean that the Cato institute's number is closer to the truth" exactly. How inflation is measured in developed economy has a lot of subjectivity e.g. what should or shouldn't be considered. To say "we have the real truth" about Zimbabwe is, plainly, bullshit. At least the IMF is honest about the uncertainty. The other piece is that it really makes no practical difference if the Zimbabwe dollar you held is worth Z$0.0000000002 or if it's really worth Z$0.00000000021231. Both are nominally zero. – JimmyJames May 14 '19 at 13:51
@reirab Your notion of empiricism is clearly different from mine. – JimmyJames May 14 '19 at 14:36
@JimmyJames I think in the end we're arguing different points here. Of course, 89 Sextillion Percent is not "the correct number", as the headline claims. On the other hand, at least on a casual scan I find no indication at all in the IMF's report on how they arrived at their number other than "IMF estimates" (if I read the table from the Cato Institute correctly, the national bank of Zimbabwe didn't publish numbers for September), and I understand how scientists could be angry if their number, which it took effort to come up with, and of which they have described in detail how they arrived... – sgf May 14 '19 at 14:36
... at it, if the Economist just cites some number the IMF whipped up for a meeting with Zimbabwe officials - that number doesn't even pretend to be scientific. What set me off on all of this is your claim that this is about ideology. But if I were a scientist and my work wasn't cited rather than some corporate body's random number, I'd be angry quite irrespective of ideology, – sgf May 14 '19 at 14:38
@sgf "I think in the end we're arguing different points here." It seems so. Since you are responding to my answer, I would expect you would be addressing my point. "Of course, 89 Sextillion Percent is not "the correct number", as the headline claims." is definitely part of my point. I'm not clear on what your point is. – JimmyJames May 14 '19 at 14:39
1

@JimmyJames " Essentially it says: "the Economist is wrong, look at this table created by us, see?"" That's not what the article is saying, and you're misrepresenting it. You're tunnel visioning on the headline, without even stating that in your answer, making it misleading. If nothing else, my comments offer an opportunity to you to improve your answer, by showing you how I may have misunderstood it to target the body of the article, rather than it's headline. That's precisely what comments are for, so I think it's safe for you to unmount your high horse. – sgf May 14 '19 at 14:41
@JimmyJames Your point, from the way I understood your answer, is: "No-one knows the real number, so they're trying to discredit the Economist for ideological reasons when they pretend their number is better." But that's a non-sequitur. They seem to have good reasons for preferring their guess: They wrote a paper that explains their number, and the IMF might as well have made it up. You're attributing to malice what can be attributed to common sense. That calling their number "the correct number" is, obviously, going too far, is irrelevant, especially since your answer doesn't talk about it. – sgf May 14 '19 at 14:48
@sgf From the body of the article: "For accurate estimates of Zimbabwe’s fantastic hyperinflation that are used in the professional literature – estimates that are reliable and replicable" and then the author sites his own study. – JimmyJames May 14 '19 at 14:48
@JimmyJames Are you actually proposing people shouldn't cite their own papers when they're pertinent? – sgf May 14 '19 at 14:49
@sgf The whole point of this article is to point out that the Economist is "publishing bad numbers" no? If it's something else, please explain it to me. – JimmyJames May 14 '19 at 14:50
@sgf " That calling their number "the correct number" is, obviously, going too far, is irrelevant, especially since your answer doesn't talk about it." That makes no sense. That is the crux of my argument. How can it be irrelevant? – JimmyJames May 14 '19 at 14:52
@JimmyJames I honestly have no idea what the article is about, apart from the author being salty that their own number isn't being published. It's an article that's filled with hyperbolic language (all the poor Economist is saying is "some estimates are..." and leaving out the author of the article) and tries to prove a very general point by a very insignificant issue about numbers that - as you rightly point out - don't really matter. My only gripe with your answer is that you seem to imply ideological bias where academic vanity is a much more likely explanation. – sgf May 14 '19 at 14:53
@JimmyJames If this is the crux of your argument, you might want to state that that's the statement you're talking about - it only occurs in the headline of the article, and it's easy to forget the headline when you've read the body. – sgf May 14 '19 at 14:54
@JimmyJames Because there's really a difference between saying "My guess is better and theirs is worse" and saying "My guess is right and theirs is wrong." You seem to be saying that they're wrongly doing the first, but from your comments I get the impression that you're just pointing out that they're wrongly doing the second. – sgf May 14 '19 at 14:55
@sgf It's also in the body, right before the table as I just quoted... I'd be more inclined to believe that this is just 'academic vanity' if is weren't for this: "It turns out that The Economist is a serial propagator of inaccurate IMF figures." [emphasis mine] – JimmyJames May 14 '19 at 14:55
Surely there's a difference between claiming that an estimate is "correct" (as they do in the title), and claiming that it is reliable, replicable, or accurate (as they do in the body). I don't see why the author of an estimate shouldn't claim all of the latter for their estimate if they did their work well. – sgf May 14 '19 at 14:57
@sgf I should remove the word 'guess'. It's not what I really mean. I'll edit it. – JimmyJames May 14 '19 at 14:58
@JimmyJames I'll do one last comment and then I'll leave you in peace :) My gripe is with this: "In other words, they are saying their guess an imprecise figure is better than someone else's." From that you argue that they're wrong. But there can be very good reasons to assume that your guess is better than someone else's. So while your conclusion may very well be true, it doesn't (to me at least) follow from what you've written. The argument may be complete in your head ("in their case, the guess isn't better because..."), but it isn't in your answer as far as I can tell. – sgf May 14 '19 at 15:00
@sgf I removed the political commentary. It's what "I got" out of it when I read it but it's drawn largely from experience and minimally content of the article. – JimmyJames May 14 '19 at 15:12
@JimmyJames What is your notion of empiricism? Note that proof and empiricism are two quite different things that really aren't even closely related (one is deductive or exhaustively inductive, while the other is inexhaustively inductive.) To say that there's no empirical way to determine something is to say you can't look at the evidence and draw meaningful inductive conclusions from it at all. That is generally false in regard to significant market movements. Saying that there's no way to prove the cause of such a movement would be more correct, though (assuming you can't just ask.) – reirab May 14 '19 at 15:26
@reirab I'll defer to wikipedia: "... must be tested against observations of the natural world rather than resting solely on a priori reasoning, intuition, or revelation." So if your argument is that "it's possible" for someone to do 'experiments' on markets and news and show some sort of relationship, I guess I'll grant you that. What I really meant is that it's not conceivably possible that this is what the financial news is derived from. I'm really taking it on the chin with my loose language here. – JimmyJames May 14 '19 at 15:37

score 2 · Answer 3 · answered May 13 '19 at 16:22

My observations is that it's worse. My broker's computer feeds me news articles on financial markets, predictive indications in various industries, etc. They are 100% bad data for investment.

This bears repeating, not one of the articles supposedly selected specifically for me was of any use at all. I could have made more money taking the opposite action as would be encouraged by the financial news I received. The only reason I did not do so is I do not dabble in shorting stock.

I have a plan, and I know where I need to be in 20 years, and I'm having no trouble getting there, and my plan doesn't involving reacting to the news in the obvious ways. I look for when the market moves and the predicted benefit from the general headlines by my own analysis are in opposition to each other. Then I profit. Obviously to do this, I must understand the industries I invest in.

Is 95% of what you read in the financial press “either wrong or irrelevant?”

3 Answers3