Where Is Germany's Gold?

That is a good argument. However, I suspect that their is a subtler form of the gold bugs' argument that has more validity but they're not phrasing (or haven't even thought about): the $500 has real, serious value, but it doesn't clearly have $500 value.

What's the value of the shirt? That depends on a lot of things that aren't known to you, and might not be known to them: how long will it last? how versatile is it? what would it cost to find an equivalent shirt, or to resell it? etc. These things are also ambiguous at a clothing store, although less so. But what is clear to both you and the goldbug is that it's clearly and obviously worth much less than $500: you're willing to estimate a rough value, even though neither of you can't put an exact number on it.

The goldbug ought to be complaining that everyone looks at a five-hundred-dollar bill and thinks that it's clearly and obviously worth exactly $500.00. That is tied into a lot of things, including, like the shirt, how long it will last and what it can be exchanged for, and neither of you know these things for sure.

With our worldview and assumptions, you and I might conclude that the bill is definitely worth at least $495: the chance of government collapse or a run on the banks, even over the next several years, is much less than one percent, and we can expect to use the bill before inflation devalues it by one percent. The goldbug, with their worldview and assumptions, might conclude that the value is fuzzier and possibly as low as $400: maybe there's a 20% chance of a run on the banks in the next few years. That still makes it worth obviously more than the shirt.

So the goldbug shouldn't be arguing that the money is unbacked and valueless, and you're right that things (including bills) are worth what people think they're worth. The goldbug should be arguing that we as a culture are overvaluing the bill. (And normal people do make these calculations: ask anyone who travels frequently between two countries with different currencies whether they prefer to keep cash in one currency or the other. Even over such a small value, there's a good chance they'll have an opinion.)

Reading between the lines of this article, there's an argument to be made that the world's large banks are not 100% trustworthy, as we'd hope. Maybe they're 95% trustworthy, but we should be skeptical of the claim that they're "beyond reproach", and simple human error, let alone malice, can cause gold to be misaccounted. And if they're misaccounting mere bars of gold, what else are they misaccounting?

And we all know that "mainstream economists" are a good source for unbiased opinions.

Sure. Just like "mainstream climate scientists", or "mainstream doctors", amirite?

But that would then imply that the five-hundred-dollar bill has value, because it can be exchanged reliably and easily for value.

The goldbugs, as far as I know, don't favor gold because it's shiny; they favor it because it's physical and it's historically been popular in so many cultures as a medium of exchange for things of real, inherent value (food, labor, housing, etc.). I think a definition of "value" that roots gold as having value because it's gold would be so self-referential as to be useless.

The offsite copy involves trustworthy people, and unlike with data, there's no client-side encryption for gold.

If you do believe that the NY Fed, the Bank of England, and the Bank of France are "above reproach," then your argument is absolutely correct. But if you can't be sure of that—and this article is definitely attempting to cast doubt on that—then there's the usual argument about self-hosting. Amazon's probably a better sysadmin than most people are, but you're the only sysadmin that you have meaningful control over the competence of.