Hacker News Reader

Comment by nordsieck | Hacker News Reader

Lots of people predict lots of catastrophes. The trick is identifying the correct ones ahead of time. This is much more difficult than you are implying.

View on HN

I think it's still very simple. We can view it through the concept of discount rate [0], basically people don't value a future event as much as a present one.

The world ending in 50 years is not worth making drastic efforts now, at least for most people. If your prediction is too costly on the short term, as accurate as it can be most people won't act on it.

[0]https://en.m.wikipedia.org/wiki/Intertemporal_choice

I suspect this is in part a side-effect of the Anglo-American (maturity-mismatched) banking system, which keeps long-term interest rates artificially low, leading to higher time preference.

the discount rate applies to utility, which may or may not have any relation to the discount rate of money.

Wouldn't a lower interest rate cause us to value the future more? (Or you mean, the natural interest rate should be higher?)

Lower interest rates increase your relative propensity to spend by lowering the opportunity cost of consumption. For example, if you have $10K and can get only a 1% annual return, you might as well take that trip to Vegas. If you could get a 15% return, though, you might rather stay home and rake in the interest.

OK, you basically have to compare your own internal discounting rate of consumption with the public rate of interest offered on savings.

If your own internal rate is higher than the public one, then indeed you should borrow to consume (or put off saving).

Companies only care about the public rate---if they have a higher rate of reliable return internally, they can keep borrowing money to invest until there's an equilibrium. A low interest rate makes companies more forward looking.

Since interest rates are a man-made concept, one must wonder what a "natural" rate of interest actually is.

Words like "Artificial vs. natural" attempt to ascribe illegitimacy to what is mostly a policy difference.

I think the parent means something else. You can find people predicting all kinds of disasters for the next 5 years and maybe some of them are right. But it's very hard for governments to know which one are right. Especially considering some are of the form "there's more than 50% possibility something bad will happen". In this case even if the bad thing doesn't happen in 5 years, you are still not sure if you should disregard the warning.

The problem is made worse by the fact that your confidence in predictions also generally diminishes quickly as you go forward in time.

There is also something called the "critic's lament" that goes like this:

If you describe a situation in too dire terms, people will conclude nothing can be done about it and do nothing. On the other hand, if you describe a situation in too rosy terms, people will conclude it will work itself out... and do nothing.

In this case that observation is unwarranted. There are some disasters which are difficult to predict and uncertain. There are others which are blindingly obvious or inescapable. Warning people that if a large hurricane hit a particular heavily populated city would be disastrous is not exactly extremely insightful when that city happens to live in one of the most heavily hurricane prone regions of the entire planet. As it turned out, Katrina was not even the worst case scenario, as it was not a Cat-5 when it hit New Orleans. Had it been, things would have been vastly worse. Similarly, a Cascadia subduction zone earthquake is not a question of whether it will happen but when. It absolutely will happen, and on a timescale similar to the age of the United States. That demands prudent planning and mitigation efforts.

If you live in Japan you can't merely shrug and say "oh, who can predict when an earthquake will come? nobody knows", no you live in an earthquake prone area, you need to plan. Similar if you live in tornado alley in the US you can't shrug and say "nobody can predict a tornado". You need to prepare.

In this case, it's not at all difficult.

Reply on news.ycombinator.com