This is true, but gold at least has a several-thousand-year history of use as a currency, and as a stable source of value.
The US dollar that we use today has a history going back to exactly 1971 and the end of Bretton Woods.
I think people get a little too worked up over the evils of "fiat currencies," but they do have a point that from a historical perspective they are basically an experiment. It's entirely possible that people may look back from a few centuries hence and look at floating national currencies (which are essentially backed only by the backing government's power to tax a particular economy) and think "what a really terrible idea." The very power that a fiat currency gives a government over its money supply might turn out to be more temptation than even the most honest government can withstand in the long run. We don't know, because there just hasn't been enough time to tell.
There's two orders of magnitude difference between the amount of time gold has been in use and the amount of time the current un-pegged dollar has been around. While it's true they're both given value by what other people will trade for them, it's a bit disingenuous to say they're the same.
The total amount of all gold ever mined is less than a tennis court cubed. In the last 2000 years, gold has been the single commodity that has always held value, particularly in times of disruption and panic. It may be volatile, but it never becomes worthless, is highly liquid (perhaps the most liquid asset around), is mobile, very durable, and stores a great deal of value in a small volume. What other assets can you say share those traits? In the 1900s alone, there have been 20+ times of complete collapse of paper currencies: http://en.wikipedia.org/wiki/Hyperinflation
While past performance isn't always an indication of future behavior, some trends are worth paying attention to.
If you want to scoff at something, let's start with diamonds which can be _manufactured_ to a higher quality, at increasingly lower costs. I mean, we are talking about _carbon_ there!
The amount of gold available in the world is relatively stable; there's not much of it and never has been.
The amount of "stuff" -- things of value which we wish to obtain through economic transactions -- is not stable and has not been for quite some time; it has, rather, been expanding at a rapid pace (consider, for example, the number of profitable industries today producing durable goods which did not exist even a half-century ago).
This represents a problem: if the total amount of currency available for use in transactions does not maintain rough parity with the total amount of goods and services available to be exchanged in those transactions, then the imbalance will lead to wild price fluctuations.
Since this situation is isomorphic to inflation (the only difference is which side of the currency/goods equation harbors the imbalance), the magnitude of the fluctuations will be the same as in an equivalent inflationary situation.
Thus, your point would be a strong one if only you were arguing for gold as an investment (since, in an imbalance of the type described above, the market value of gold would increase). As an argument for gold as a currency, it holds no water and in fact tends to undermine that position. In the present age, gold simply cannot be the foundation of a stable currency.
In the case of a fixed amount of currency in an expanding economy, or if the economy expands at a rate greater than the amount of currency expands, you simply have deflation.
I'm not convinced this is nearly as bad a thing as some growth-uber-alles economists seem to think it is; it would encourage savings and cautious investment while preventing bubble formation. It would slow growth, but endless growth isn't sustainable anyway. (And increasing the money supply too fast, which there's always a lot of temptation to do, fuels highly unsustainable faux-growth.)
Also, although the amount of gold on and in the planet is fixed, the amount usable as a currency increases over time as it's extracted. At least in the short and medium term (probably at least the next few centuries) you would be able to have mild to moderate economic growth without deflation as the gold supply increased.
The best argument I've heard against gold-based currencies is that is mis-allocates resources: it creates a huge incentive to go around digging up gold, which is an activity serving no real useful purpose. When someone spends time and energy mining gold, they're not doing anything else with it; it's basically a dead-weight loss. Still, compared with currencies that can be created arbitrarily it might be an acceptable tradeoff.
I take issues with your qualification that it's "simply" deflation.
Inflation is taken to be a bad thing since, if the price (in whatever currency we're using) of goods and services suddenly increases by, say, 1000% in a short time, people who purchase those goods and services are screwed, which also means people who sell them are eventually screwed (since they have no market in which to sell their goods and services).
But deflation is at least equally bad: if I _sell_ goods or services, and the price (again in whatever currency we're using) drops 1000% in a short time, then I and everyone else who sells goods and services is screwed, which means people who would buy goods and services are screwed (since everybody who was selling them just went out of business).
This is why gold can be a good investment in the modern age, but not a good currency; any time a new development results in rapid economic expansion, it would be accompanied by wild fluctuations in the (gold) currency, tending toward crippling deflation.
It's also why I'm harping on the fact that money _as money_ must by definition have no intrinsic value. Its job is not to have value. Its job is to stand in for things of value so that we don't have to use insanely complex barter schemes.
And since gold is, for reasons surpassing understanding, held to have value far beyond anything attributable to the actual substance (which is "valuable" in itself only as, for example, part of electrical components), gold is an absolutely terrible basis for a currency in this modern world.
With respect to your second paragraph, the economics profession has far from sufficient data for anyone to really hold a scientifically derived opinion on the matter.
The purpose of currency in an economy is (increasingly) seeming to include imposing a tax on using money as a store of value. See Mankiw's recent remarks on negative interest rates, which are all consistent with the mainstream monetarist theories that underly our use of fiat currency, etc.
A trend that I am noticing is the idea of credit as a "right" just as people consider healthcare a right and perhaps home ownership. We have already seen a bit of politicization of the Fed. I sincerely hope that this is all nonsense, but the danger is that the most powerful interests would actually stand to gain from such a regime. At present, banks (who are the biggest contributors to both parties) would gain as would various other industries who (by virtue of various other government imposed incentives) are popular ways that people attempt to purchase stored value (houses, 401Ks, etc.).
I should have called out the one element that I was responding to,
--SNIP--
As "money", both are valueless in and of themselves, and merely stand in for things of actual value (goods and services).
--SNIP--
But, in hindsight, I realize that your statement, wasn't suggesting that gold had no value, but that its value was derived from those things that it stood for - which is, to some degree, orthogonal to the fact that it historically has reliably, and consistently, stood for things of value.
Right, it's not a matter of "intrinsic value" - the value of a thing is determined by supply and demand. However, it's a hell of a lot easier to increase the supply of dollars than the supply of gold.
