Compound interest can help save for the future, or it can bankrupt the world | Hacker News Reader

Compound interest can help save for the future, or it can bankrupt the world

laphamsquarterly.org

55 pointsThevet31 comments

This article seems to be based on a fundamental misunderstanding of economics.

A trust can't grow faster than the value of its investments, and on average, trusts won't grow faster than the economy as a whole. So there's very little risk that a trust will grow to encompass a significant portion of total wealth in a way that could threaten an economy.

Science fiction stories that show someone waking up to find their small balance has grown into a fortune are just that, fiction. Anyone who's kept money in a bank knows that the interest doesn't even keep up with inflation.

I'm confused by this comment. That sounds like a savings account you're talking about. But a T-Bill grows at inflation + (small percentage, though probably still less than GDP growth). And a stock market fund grows faster still (and in general, the stock market has shown sustained growth when measured over decades--though one might question that after the last 15 years). Given an index fund invested in 1929, you would have growth beyond what the economy experienced during that period, right?

There are questions about how you'd manage a multi-trillion dollar fund: you'd eventually reach the point where you can't just use index funds. But other than that, your comment doesn't seem accurate.

Edit: What is true is that there is a limit to what your investment could become. For obvious reasons, it can't be more than the wealth of the world, and there would be asymptotic behavior at even less than that.

I'm only referring to a bank account in the final two sentences. There is a common trope in time travel stories where a person finds that their small bank account has become a fortune. The linked article seems to think that that's not just reasonable, but likely. But it's obvious nonsense.

Yes, a T-bill grows faster than inflation, but slower than the economy as a whole. So you're going to get a smaller and smaller piece of the total economy--completely at odds with the premise of the article.

The rest of my comment is valid regardless of the assets the trust holds. Certainly, someone can get a bigger slice of the economic pie by investing wisely (or luckily), but there's no particular reason to fear that a long-lived trust will do so, and plenty of reasons to think they won't. Look at successful companies from 100 years ago. Most of them don't even exist any more. Why would we think the management of a long-trust would be better at navigating the future economy than that of a major company?

That trope has a dual premise. First, compound interest is in play. Second, the beneficiary is not drawing on the account for living expenses.

If you can put $10 in an account at 3% interest and wait 200 years, you can take $3693.56 out. The key is not needing that money at all over a 200 year time span. The interest only accumulates on money that stays in the account.

Rich folks can afford to let money sit and gather interest. Poor folks spend all that they earn, if not more.

As a thought experiment, take two identical trust funds. Put $10M in one, and $20M in the other. Both grow at 4% per year. At the end of each year, the beneficiaries may withdraw up to 3% of the fund. The beneficiaries have similar tastes, and initially withdraw $250k. Each subsequent year, they draw 3% more, unless they hit the cap.

How many years before the larger fund grows from double to triple the size of the smaller? 25. It is four times the size after 42 years. It is 10x the size after 112 years. It takes 14 years for the less endowed beneficiary to hit the withdrawal cap, and 128 years for the other to reach it, at 10.64 times the size, spending 10.64 times as much per year, forevermore.

If your expenses grow more slowly than your investments, you will grow ever richer. If your spending grows faster than your income, such as for everyone in the US earning a wage or salary since 1970, you grow poorer.

> If you can put $10 in an account at 3% interest and wait 200 years, you can take $3693.56 out.

Yes, but how much is that $3693.56 worth in 200-years-ago dollars? Chances are, the number is less than $10.

In 200 years, other things could also happen.

- US/Federal Reserve Dollars could be obsoleted in favor of a different currency. Your account might automatically be converted to a non-interest bearing account in a different unit of account.

- Your account could be flagged as inactive and plundered by the banking institution or the governing authority.

- The currency could be inflated much faster than your fixed rate of return. Your $3693.56 could buy one ramen noodle--not one packet of noodles, one noodle.

- The banking institution could cease to exist. If there is a successor institution for your account at all, you might not know what it is.

But the point is that if you just take your $10 bill into the future with you, you would only have one worthless 200-year-old $10 note, of interest only to curators and collectors. You wouldn't have the not-even-$10-worth of era-appropriate currency.

Cf. Berkshire Hathaway.

> For obvious reasons, it can't be more than the wealth of the world

I think your reductio ad absurdum carries more weight than you give it. Real risk-free interest can't on average over a long period be greater than economic growth. If capital is always rewarded more than the value it creates (economic growth is the meaure of the total value created) then where does the extra value come from? It has to come from labor. Labor will not, in the end, work for nothing (obligatory Pikkety reference should be inserted here.)

The sum of all capital in the world can only take a share of economic growth, some share must go to labor. That means that if there is an investment that returns more than economic growth, there must be an investment that returns less. As the pool of money being invested grows larger, it has to take part in both.

It has to come from labor

Or information asymmetry?

But then it's coming at the expense of some other investor (unless I'm misunderstanding.) No value is created. This might work if the amount being invested is small, but not as it grows larger.

Yes, at the expense of someone else.

It can still work as the amount invested grows, if you are in a position with a reliable feed of such information. See Plunkitt's "honest graft".

My point was to capital as a whole (that is, all of the capital.) When I say "get large" I meant as a large percentage of all capital. Even multi-billion dollar funds have this problem, and that's just a tiny fraction of all capital.

I agree that it can't even get near 100% of economic value (though I don't have a clue how to make 'near' precise). I was being handwavy about how much, and could've been clearer.

One of the things that investors soon learn is that achieving any sort of compound growth over a long time frame is virtually impossible.

>you would have growth beyond what the economy experienced during that period, right?

Are you looking at US growth, or global growth?

In the U.S. average inflation over the last century has been 3.2%, while the average interest rate of 10 year treasuries has been 4.9%, so you'd been able to grow your fortunes 5.4 times in real dollars, at a rate of 1.7% per year. That rate is smaller than the average rate of GDP growth, however.

So, after investing your money for a century, you'd be a lot richer than in 1915, but you'd own a smaller slice of the pie.

Capital's share of the economy definitely can grow faster than than the economy as a whole. There's a whole book about this:

http://en.wikipedia.org/wiki/Capital_in_the_Twenty-First_Cen...

But it cannot surpass the total value of an economy, and if you started with a tiny fraction of all capital at the start, your portion of final capital will likely be around that same small portion of all capital.

Thus the premise stands. Compound interest is not a parasite eating value from other items. It is only a return on investment, which everyone is competing for.

It could happen, hypothetically, if you repeatedly picked winners and losers and in that way steadily grew your share of the economy. But it's quite infeasible.

Trust funds can grow faster than the economy depending on the assets. I cannot remember where I read it, but warren buffet said stocks return on average about 7%. 2% for inflation, 3-4 % for economic growth, and the rest from dividends. (It's not much more, but dividends still add some value)

I suspect the concerns about how large a trust might grow are just camouflage for the people trying to get at the money early.

None of the trusts ever actually grew to be that significant when compared to the wealth of some living persons and dynasties -- the Rockefellers and the Kochs and the Waltons. The set of people terribly concerned about the undo political influence of those running a perpetual trust seems disjoint from the people concerned about the undo political influence of the living rich.

You can not simply invest your money at "compound interest" for an arbitrary number of years. Market interest rates change over time. And even if you would manage to get some institution to guarantee you interest x for n years, that institution might go bankrupt before you get your money back.

Some long running investments exist, like governments still paying off debts from World War 1. But you can not count on such deals being available when you need them.

I think Andreas Eschbach handled it pretty well in his excellent novel One Trillion Dollars. It also has the premise of a heritage that compounds for a very long time. But it explains that there was a family whose sole purpose was to manage the fund for centuries until the heir was ready. (The book is not so much abut the wonders of compound interest as about the value of money).

Even if someone was infinitely rich it would merely cause hyperinflation. This article seems to have quite a bit of economics misunderstandings.

There was an econ professor that gave a guest lecture early in my college years that said inflation is merely a societal disagreement about the distribution of income between the wealthy and everyone else. He explained why this is the case very well at the time, but I can't remember the details now. If anyone else has been told this and can explain it here, it would be greatly appreciated.

IANAEconomist, but I'll try to take a stab at it. Let's assume a growing civilization, and a government that has imposed a condition that a given currency must be accepted for debts.

Deflation is the default, since currency-tokens don't multiply on their own. With deflation, the reward from "overall civilization growth" automatically go to the folks who have either actual tokens in a vault, or who are creditors to somebody who promised to get them an actual token later. (This is usually "wealthy" people.) Conversely, all the people in debt get a bad deal, where even a "zero-interest" loan is harder and harder to pay back as their debt of "one token" means more and more goods/labor.

Conversely, inflation helps the people with debts, because (all else being equal) they become easier to repay as time goes on. Because people with net-debts are usually "the poor", it follows that "inflation is pro-poor."

Arguing over who gets the "new tokens" being minted to combat deflation is actually a separate (albeit intensely related) debate, and that's where you hear people complaining about the Federal Reserve and stuff like that.

He was most likely referring to the view that bargaining over the share of company revenues spent on wages is often a driver of inflation. In brief: Employees feel that their company's growth should be reflected in higher wage incomes for them (especially if they're worried about higher prices in future) and so seek to negotiate higher wages, to the extent labour market conditions permit them to. Business owners feel the benefits of the growth should instead accumulate to them and therefore seek to protect their profit margins by increasing prices (especially if they're worried their suppliers might put the prices up in future), to the extent conditions in their market permit.

I think this was similar to the argument he made.

This could not happen in the U.S. because it would violate the Rule against Perpetuities.

http://en.wikipedia.org/wiki/Rule_against_perpetuities

However, compound interest is a very powerful tool.

http://www.daveramsey.com/article/how-teens-can-become-milli...

>This could not happen in the U.S. because it would violate the Rule against Perpetuities.

From the same link, not all states follow that. Many others adopt a "wait and see" rule, which may change one way or the other over the relevant time period.

So it's not quite true to claim this could not happen in the US.

"Give me control of a nation's money supply, and I care not who makes its laws." --Rothschild in 1744.

This article is at least as old as 2011, since there was an MR discussion of it then: http://marginalrevolution.com/marginalrevolution/2011/09/the...

Reply on news.ycombinator.com