31

I understand that the stock market is important. Even when people don't directly invest in it, they might still have pension savings which will be invested in the stock market. Returns over a 20–50 year period thus make a real difference to standards of living upon retirement.

But why should I care if the stock markets drop 10% on one day and rise 5% the day after? It's a symptom of economic uncertainty due to current events, which lead to real economic problems for some, but we already knew that, with millions of tourism jobs at risk, transportation companies (temporarily) laying off people, schools closing making it harder for some people to earn a living, and global trade affecting supply lines, all real effects affecting real people. I'm also aware of high frequency traders earning a lot of money because they receive news 5 milliseconds before others, but that is on a millisecond level, not a daily level. Yet large parts of the press seem quite focussed on daily (or even sub-daily) stock market fluctuations. Why?

3
  • 30
    You are correct that the news habitually reports "markets are behaving normally" as if that were news. The news also habitually reports stuff like "a regularly scheduled sporting event ended in a win for one of the participants" like that was news. The conclusion you should draw is that a significant fraction of "news" is pointless filler because they are required to use up a certain amount of time, and actually reporting and analyzing the issues that genuinely affect people is (1) expensive, and (2) contrary to the interests of the people who own the news. Mar 13 '20 at 21:27
  • 3
    You should not. Although it's nice when you can pick up some stocks on sale... Mar 14 '20 at 2:08
  • 5
    @EricLippert I disagree with the sporting event example. It might be pointless to you (and in this I actually agree with you) but that does not mean it is not newsworthy.
    – jcm
    Mar 14 '20 at 4:41
39

Short-term fluctuations are more significant than they might seem because:

  1. They are not as small in relation to long-term returns as one might expect. For Brownian motion (random walk) the typical move over a given time interval scales only weakly, with the square root of the time interval. So a really bad day can wipe out a whole year's return.

  2. They are not transient, i.e., stock prices are not mean-reverting. Rather, stock prices behave as a martingale where today's price determines the expectation value of all future prices. Thus, if you are retired and living off a stock portfolio, when the market drops 10% (no matter in a day, a week, or a year), your expected standard of living for the rest of your life just dropped 10%. (That doesn't mean you must immediately reduce your living expenses by 10%, of course, but you would do so if you want to keep your risk of running out of money unchanged.)

9
  • 9
    I find it interesting and counterintuitive that they are not mean-reverting, I should read up on that.
    – gerrit
    Mar 13 '20 at 14:54
  • 10
    It may seem less counterintuitive when you note that if they were mean-reverting, it would be free money -- buy dips, sell rallies, and easily beat buy-and-hold.
    – nanoman
    Mar 13 '20 at 17:15
  • 2
    "Rather, stock prices behave as a martingale where today's price determines the expectation value of all future prices." Evidence that this is actually accurate in practice?
    – usul
    Mar 14 '20 at 13:59
  • 4
    @TannerSwett I'm not convinced it's as simple as that. (1) your logic is only true for risk-neutral traders. (2) stocks are sometimes very correlated as in the recent crash, which seems to mean (3) even the largest institutional traders cannot be risk-neutral inside the market, plus (4) they have a complicated risk structure due to bonuses/incentives for individual decisionmakers and the fallback of bankruptcy or a bailout. ... So that's why I asked if there is evidence.
    – usul
    Mar 15 '20 at 5:11
  • 7
    Time horizons and discounting are also an issue with the logic - see also "the market can stay irrational longer than you can stay solvent..." e.g. maybe you could have spent the last 3 years betting against the market, and nobody would have invested with your firm and you'd go out of business, before finding out that you were right.
    – usul
    Mar 15 '20 at 5:15
9

For long term investors, daily stock market fluctuation is noise. However, the market falling 25% in less than two months has a much greater implications than daily noise or the affects on some specific groups of people (tourism workers, transportation companies, etc.).

The consequences of a deep global recession affects every industry. Apart from job loss, many will lose their homes and be unable to afford basic necessities. Just look back at the 2008 GFC and the observe the hard consequences of that.

FWIW, high frequency trading have nothing to do with this. They're 'arbing' their millisecond speed advantage regardless of whether we're in a normal or a volatile market.

4
  • 1
    Of course a global recession matters, but is a 25% drop in two months a cause or a symptom of a recession?
    – gerrit
    Mar 13 '20 at 14:51
  • 1
    The market is forward looking. A 25% drop in two months is an expectation of a recession. Mar 13 '20 at 15:26
  • 4
    But can that be a self-fulfilling prophecy?
    – gerrit
    Mar 13 '20 at 17:09
  • I wouldn't call it self-fulfilling but to some it extent it is contributory. Mar 13 '20 at 17:45
5

Like earthquakes, it depends how large the shift is.

Minor changes don't generally matter

The stock markets tend to bounce around inside a small window. I'll pick 5% to be safe (I rarely hear of the stock markets over a 5% drop). So if the market goes up or down 5% in a week's time, that's not a very big deal. Yes, that's impactful to someone, but it's a minor gain or loss. Here's an NYSE composite for the last year

NYSE Composite

Note that there's ups and downs, but the gradual trend is upward prior to the last week. In most cases, this is where the market tends to be. Worrying over these smaller fluctuations isn't worth it. Markets tend to go up over time. If you have enough distributed risk (such as mutual funds), you might not feel much, if at all.

Major drops signify something is wrong

At yesterday's close, the market was down about 20% from its peak just a week or so ago. That's not a fluctuation, that's a correction

In investing, a correction is a decline of 10% or more in the price of a security from its most recent peak. Corrections can happen to individual assets, like an individual stock or bond, or to an index measuring a group of assets.

The catch here is why you're investing. If you have a 401k and 20 years to retirement, worrying about a correction today is silly. The market will recover eventually. Long term goals should not be altered over corrections.

If you have money in the market because it's being used for short-term needs (i.e. a medical trust or a personal use account), this is a lot more painful because you still have to sell, but that stock is worth a lot less.

Worse, it might take a lot of sliding to get back to where the market can go back up. In 2008, the cause of the slide was the spike in defaults in the US mortgage market after the real estate bubble burst. Today, it's the coronavirus shuttering entire industries for an indefinite period. We might not have hit bottom either. As such, being concerned about investments here isn't a bad thing at all. The key is not to panic.

2
  • Your chart is only 1 year. Saying the stock market bounces around in a 5% window is crazy. This is recency bias. Average returns (let’s use 7%) are not the normal. One year you could get 22%, next year you could get -14%, it’s no big deal, it all comes out in the wash.
    – Adam
    Mar 13 '20 at 22:02
  • @Adam Are you saying the stock market has never lost 5% in the course a week? Such things are typically mentioned as footnotes on the news. 5% in a day would be noteworthy.
    – Machavity
    Mar 13 '20 at 22:21
1

You are absolutely right. If you are long-term investor, you mostly do not care about day2day changes. The least-sized scale you trade is basically 1 month candle or 20% price change. You are also would be working mostly with fundamental information, ignoring stock price history at all.

BUT. Even while you do not trade on low scales does not mean that you should ignore technical information. Check out indicators such as volatility, volume-based indexes etc. because they could possibly provide information for you to understand fundamental issues of your instruments.

1

There are some great answers already like nanoman's and machavity's, but there's an additional worry, especially if you own individual stocks:

stocks can drop to zero.

For instance down 25% back up 28% for a 3% net only works, crucially, if the drop didn't wipe you out and you are still in the game to bounce back.

This worry is in addition to all the other worries like the knock-on halo effects. As to why you should worry about this on the timescale you reference, well, there's a saying in stocks that a bull comes up on the stairs but a bear goes out the window. Stocks fall faster than they build up/recover. As the other answers point out this is a problem if you're retired and not still contributing earned income, but it's a bigger problem if your assets become completely worthless, and that can happen in shockingly short periods of time.

Owning stock indirectly shields you from much of this worry, but really big shocks can cause the bankruptcy of whole firms (including the one running your mutual fund).

0
But why should I care if the stock markets drop 10% on one day and rise 5% the day after?

Because if you are actively investing into stocks, that opens you to a very great purchase window. Think like this: if there's something you have for long wanted to buy and you see -10% discount on it, will you buy it? Most probably.

Similarly, if you are planning to buy stocks and there's a discount, you should take advantage of the opportunity and do the purchase for the discounted price.

I also disagree with nanoman about mean-reversion property of stocks. Stocks do have a mean reversion property. Think like this: if you own a stock with 1 USD dividend that costs 10 USD, and then its price will drop to 1.5 USD, do you expect its price to stay at 1.5 USD (+ natural growth) in the future? You don't.

To make that more extreme: if the dividend is 1 USD and the stock price is 0.1 USD, do you think its price will stay at 0.1 USD forever? No it doesn't.

1
  • It doesn't. The company might declare insolvency.
    – gerrit
    Mar 16 '20 at 17:41

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.