[tcbawo]

Gold was historically uncorrelated to other asset classes, so there was a portfolio theory reason to allocate in gold. In times of uncertainty and fear, people would flee to gold. Afaik, this effect has been muted in recent years. I don’t know if it has been fully explained, but it’s likely due to the highly available and interconnected investment products of today. Bitcoin has been shown to have a high beta with credit and other investment assets. So from a portfolio theory perspective, it is functionally more like a leveraged ETF than a hedge. This doesn’t mean that neither deserves a place in a modern portfolio. As they say, past returns are no guarantee of future results.

[rsynnott]

It’s arguably a _bit_ of a myth that people flee to gold in times of uncertainty. It’s something that many people believe happens, and it does seem to happen to _some_ extent, but not to the extent that people believe. Adjusted for inflation, gold has never reached its 1980s peak again, and may never do so, despite a number of greater economic and social shocks since that peak.

[DennisP]

So it bubbled and crashed forty years ago. In the past 25 years it's had decent returns, and in particular:

2007: 31.59%, 2008: 3.41%, 2009: 27.63%, 2010: 27.74% [1]

Owning too much gold is terrible, but a modest allocation to gold in a portfolio of mostly stocks, which you rebalance every year or two, works out pretty well. [2]

[1] https://www.macrotrends.net/1333/historical-gold-prices-100-...

[2] https://portfoliocharts.com/2021/12/16/three-secret-ingredie...

[rsynnott]

I mean, each to their own. I don’t understand how anyone can look at that chart, and look at, say, an S&P500 returns chart (or an MSCI World chart if you want some more diversity) and say “yes, gold is a good long-term investment”, but you do you.

(NB. Gold enthusiasts will sometimes try to cloud the waters by comparing gold to, say, S&P500 price only (ie pretending dividends aren’t a thing). This honestly still doesn’t leave gold looking very attractive, but it’s also dishonest. To compare like with like you want to look at S&P500 with dividends reinvested, and at that point gold just isn’t even in the running.)

[DennisP]

As I wrote above, by itself gold is a terrible investment. But as a small portion of a portfolio that you rebalance periodically, it has its benefits for improving risk-adjusted returns. See my second link for details.

But you can get the gist by observing that in 2008, the S&P500 dropped 38.49%,[1] while gold went up 3%. If you had, say, 10% of your portfolio in gold, then when you did your end-of-year rebalance you got to buy a lot of cheap stock. Gold also substantially outperformed stocks in 2007, 2009, and 2010. So I'm not convinced it's a "myth that people flee to gold in times of uncertainty." Maybe you're right that it's not as big an effect as some people believe, but it's enough to benefit your portfolio.

[1] https://www.macrotrends.net/2526/sp-500-historical-annual-re...

[tcbawo]

Portfolio theory suggests holding uncorrelated assets and rebalancing. If/when the stock market has a down year, it hopefully softens the drawdown.

[eadmund]

> gold has never reached its 1980s peak again, and may never do so

Very true. OTOH, if you bought gold one week, one month, six months, one year, five years, ten years or twenty years ago you would have made money.

I am not really a gold bug. It has a place, but it’s not the best asset ever. As others correctly point out, it is unproductive in itself. OTOH, if you have two uncorrelated assets, you can make money simply by rebalancing periodically between them. Not a lot of money, but not nothing.

Note too that cash itself is unproductive. Dollars ultimately derive their value from the fact that Americans have to pay their taxes in them, not because pieces of linen paper are useful for a lot.

[rsynnott]

> OTOH, if you bought gold one week, one month, six months, one year, five years, ten years or twenty years ago you would have made money.

Well, yes, but that's a rather unusual condition (the last time that would have been true would have been for a period in 2011, and before that for like a day or two in early 1980).

By contrast if you'd bought a broad index fund you'd have made a lot _more_ money (except for the one week example, due to Nvidia shitting the bed today). Like, a _lot_ more money.

> Note too that cash itself is unproductive.

Sure; approximately no-one will argue with you on that one. But cash isn't really the alternative to gold.

[wbl]

If all you give me is leveraged beta I can replicate that more cheaply.