Why does currency value change?

After yesterday's article about conversion between the value of
british pounds in the '70s versus british pounds today, someone sent me a link to
an article at the National Review Online, which just about had me rolling on the floor laughing. The problem is, it's dead serious.

It's written by an "engineer" named Louis Woodhill, who argues from what he calls
an "engineering viewpoint" that the whole idea of fluctuating currency value is total nonsense, and we'd all be better off if we just assigned a fixed value to our currency, and never allowed it to change.

The U.S. dollar is in a scary slide. Gold and oil are hitting record highs,
while the dollar is hitting record lows. To get how strange this all seems to an engineer
like me, imagine the following headline: "Foot Falls against Meter for Fifth Straight Day."

The accompanying article would breathlessly report that after the U.S. abandoned its "antiquated" fixed-exchange-rate system (one foot equals 0.3048 meters), our beloved foot began plunging in length. A "length trader" would predict that if the foot fell below the "psychologically important 0.2800 meter support level," it could fall as low as 0.2500 meters. But an economist would say that as long as the foot didn't fall more than 10 percent, everything would be okay.

The story would then describe the plight of a homeowner whose garage was no longer within
his lot lines. Then another economist would argue that the falling length of the foot was
actually a good thing, because it caused people to be taller, which reduced their "body mass
indexes," thus fighting obesity. The head of the U.S. Bureau of Standards would be quoted as
saying the bureau is committed to "a strong foot," although, "given that imports are longer
than exports, there is only so much we can do." The story would conclude with Paul Krugman
blaming the falling foot on "Bush's tax cuts for the rich."

What is going on with the dollar right now is every bit as ridiculous as the fictional story above. Here's how an engineer would explain the problem.

Economic transactions involve the exchange of "something" for "money." The "something" is specified in terms of number (1, 2, 3, etc.); length/area/volume ("the foot"); weight ("the pound"); and/or time ("the second"). "Money" is specified in terms of "the dollar."

The problem with this scheme is that the magnitude of our fundamental unit of market value, "the dollar," is not defined. Being undefined, the value of the dollar can change. This fact gives rise to huge economic costs and risks for which there are no offsetting benefits.

Sorry, Mr. Woodhill, but that's not how an engineer would explain the problem. It's how a pig-ignorant idiot would explain the problem. The explanation of why is beneath the fold.

It's really quite astonishing that this kind of crap still continually pops up. It's based on a fundamental ignorance of very basic economic issues. It's especially
amusing that this kind of nonsense generally seems to be printed in things like NRO - an extremely conservative, pro-free-market rag.

Why does currency fluctuate? Because the relative values of different valuables fluctuate. Peg the currency to any one, and it will stay constant relative to that
one, but it will vary dramatically compared to others. There's no way around that: it's all about supply and demand. Take things in limited supply that people want/need, and their value
will increase. The whole concept of value is intrinsically variable - value describes
a variable relation between different quantities. What Woodhill wants to do is to somehow
declare that the variable relation can no longer change. But you can't do that - you can't
simply declare that a variable will stop changing. The derivative of currency value relative to goods isn't 0, and you can't make it be.

For example, oil is currently incredibly expensive. 10 years ago it was comparatively
extremely cheap. If we'd pegged our currency to the value of gold ten years ago, then the
purchasing power - the effective value - of the dollar would be significantly less today that it would have been when it was pegged.

In response to this, America has started pumping lots of money into ethanol production. As a result, supplies of some agricultural products, like corn, have had their demand increase rather dramatically - and so their prices have risen. And that in turn is increasing the price of other things that rely on corn - like many foods. So if we pegged our currency to gold, we'd still have seen an effective change in the price of a huge swath of consumer goods, caused by the change in the value of oil relative to the value of gold.

Suppose we pegged it to oil instead - so a dollar was fixed as the price of one gallon
of refined gasoline. The dollar would be worth significantly more now than it was 10 years ago - because oil is worth so much more. But because people are paid in dollars, that would effectively make US labor dramatically more expensive, which would make it very hard for us to export goods to other countries - the production cost of American goods would be so much more expensive relative to goods produced elsewhere.

Try a different tack: suppose we set the dollar to something like a stock index: a dollar
is the combined value of 1/50th of an ounce of gold, 1/10th of a gallon of gas, 1/2 a bushel of corn, etc.

What happens then is that the dollar becomes an effective exchange medium: if oil becomes
more scarce, then its value relative to other goods should increase. But the fixed exchange
of the indexed dollar would mean that you could use it to "trade" corn for oil at a fixed
rate. How do you make the dollar work when people are gaming it like that? You need to infuse a huge amount of capital into the system to cover what people are pulling out via the exchange system.

No matter what you do, currency is going to fluctuate. The only way to stop it from
fluctuating is to stop the entire world from changing in any way. No unusual weather, no
consumable resources, no changes in population, no changes in fashion, no changes in
technology, no changes in medical care, no changes in childbirth rates - no changes of any
kind. Because the moment you allow change of any kind into the economic system, it's going
to produce changes in the relative values of different goods. And the instant that different things have shifting relative values, you cannot possible have a currency whose value is fixed relative to all of them. Pick any one, and you wind up in basically exactly the same situation that we have today with our floating exchangeable currency.

What's Woodhill's argument that you can fix the value of currency? Basically,
he differentiates between currency and capital - two different, but related, economic concepts:

Why would the Fed employ a monetary-control approach that is both indeterminate and dangerous? I believe the underlying problem is an intellectual confusion between "money" and "capital."

"Capital" is measured in terms of money (dollars), is mobilized by money, but is not money. Capital represents real economic resources. The Fed cannot create capital. All it can do is create money and use that money to commandeer capital. Unfortunately, this can cause inflation.

Once inflation gets going it tends to run away, with rapidly rising prices and escalating inflationary expectations. Ultimately this must be stopped. Unfortunately, raising the fed funds rate in order to halt inflation can cause an economy to "overshoot" into recession. Then, to fight the recession, the Fed will cut its fed funds target, thus starting the next oscillation of the business cycle.

This is fairly typical ignorant, poorly thought out rubbish. What is capital? Capital is,
to put it simply, stuff which is invested in the production of other valuable stuff.
And what is money? Money is an abstract representation of a fixed amount of stuff, which is
exchangeable for other stuff.

As Mr. Woodhill admits, capital is measured in terms of money. In fact,
capital is exchangeable for money - when you invest in a company, what you're
basically doing is exchanging a bit of currency for a bit of capital.

So how does differentiating between currency and capital accomplish anything? Frankly, it doesn't. It doesn't avoid the fundamental problem - that there is no fixed unit of value, because the relative value of different things varies. It's just handwaving nonsense. What he's effectively saying is that there needs to be two kinds of money - a floating money supply for capital, and another one for regular currency. But that solves nothing - the two will float relative to each other, and relative the values of other goods.

You can't turn a variable quantity into a fixed one.

As an aside, it's interesting to me - particularly in light of my recent career change from full-time research to software engineering - that articles like this are almost always written by engineers, who always flout their engineering knowledge as the reason why they should be believed. But when you read pretty much any article about math or science where the author touts the fact that they're an engineer, it's almost always total crackpot nonsense. I don't quite understand it.

(And for the people who've been criticizing my spelling: this article was spell-checked by emacs.)

More like this

Your engineering observation isn't limited to math; other people have observed similar engineer-related phenomena relating to writing about evolution, paranormal junk, and so on. There seems to be a (quite rare) type of engineer who confused his degree for a science degree and thinks it confers general expertise on just about any subject.

Of course, this isn't limited to engineers -- scientists often get themselves into trouble when they're writing outside their area. The Dissent from Darwin list is full of non-biologists (and includes some mathematicians, IIRC)...

But can't you see that he's right? Think of it this way:

Imagine the chaos in sports -- take baseball as an example -- if the ranking of teams was variable. What if the Yankees could be ranked lower than the Red Sox, or the Rockies higher than the Dodgers? How would fans know who to root for, which games to go to, or what over-priced paraphernalia to buy?

Why, the media would have to publish updates of the rankings *every* day. What a waste.

Now do you understand?

Read "The Creature from Jekyll Island". It explains the formation of the Federal Reserve System which is nothing but a cartel of the biggest banks.

They're the ones that took us off a gold standard promising fiscal stability when there has been no such thing. We've seen a depression and several recessions under the stewardship of the fed.

I understand the law of supply and demand, however allowing us release from the gold standard lead to ever greater government spending.

Maybe I'm misreading what you said with: "If we'd pegged our currency to the value of gold ten years ago, then the purchasing power - the effective value - of the dollar would be significantly less today that it would have been when it was pegged." but oil vs gold have tracked much closer than oil vs the dollar.

Correct me if I'm wrong but surely if there was a gold backed dollar, the quantity of oil a dollar would buy would be roughly the same?

Regards

Many people lack education in subjects they consider themselves expert enough to write about. Engineers may be over represented in this group due to their degree requirements in university. A few electives would clear the problem right up.

Hey, applying engineering principles to finance and economics gave us Long Term Capital Management, the 1987 Market Crash (brought on by portfolio insurance) and CDO underwriting. Perhaps this guy has it right?

So the National Review does not believe in markets anymore?

The problem with the gold standard is that it makes the money supply arbitrarily fixed. Under the gold standard severe booms and busts were the rule (you cannot really include the Depression under either the gold standard or the current system as Hoover tightened money even more than the Gold standard would have done naturally)

Ideally, the money supply should grow inline with the economy. The reason inflation has been contained for the past 25 years is that this has largely happened. Milton Friedman proposed mechanical expansion of 3 or 4 percent a year.

Floating exchange rates deal with supply and demand for a given currenty relative to others. Trade is balanced with capital flows. Differentials in interest rates and perceived inflation risks also significantly drive relative values of currencies.

As both an expat and a former stock broker, I feel the value of the dollar is directly related to the size of the federal deficit. It was at it's hightest point on September 10-11, the year of the World Trade Center Bombing. It's gone down ever since, especially as the size of the deficit has grown with war spending. Low interest rates in the United States keep the dollar low, too (not that we'd want to raise them now).

I don't see the dollar recovering until progress is made paying off the deficit (20-30 years?).

I really enjoyed your post.

Madame Monet (Marrakesh, Morocco)
Writing, Painting, Music, and Wine
winewriter.wordpress.com

I find it amusing (and sad) that so many people use their degrees to justify spouting all sorts of nonsense, and then accuse you of being unqualified to talk about math or creationism because you're a computer scientist and don't have a PhD in math or biology. How hard is it to realize that your qualifications to discuss a certain subject have much more to do with how much understanding you have of the subject, and how much common sense you can apply, rather than necessarily what you studied in college?

There's some irony in your spell-check comment at the bottom, because you have "flout" where you meant "flaunt". But we all know what you mean, so no big deal. Nice evisceration of a stupid argument, thanks.

By Joshua Zucker (not verified) on 13 Feb 2008 #permalink

Maybe I'm wrong about this, but I always thought of currency as like shares of stock in a country.

Just as when a company prints new shares (and gives them away), it dilutes the value of the old shares, so when a country prints new money (i.e., deflates the currency), it dilutes the value of the old money (inflation).

The international currency exchange is just the market's perception of the relative "worth" of the country's "stock".

GWB borrowing $168 Billion more to send checks to every US taxpayer will have no adverse impact on US currency. Discuss.

Bonus question: cite 3 countries that borrowed their way out of recession.

Great article. It touches on two interesting areas:

  • Most people, both those who have them and those who don't, tend to overvalue social designators of intellectual status like degrees. In other words, they believe that stereotypes are a great timesaver. People would be outraged if this were done with race or ethnicity or gender or any of a dozen other things as the basis of the stereotype, but nobody thinks twice if it's based on a degree. I've met plenty of idiots with PhDs, and I've met geniuses who barely graduated high school. Neither is the rule, of course, but there are enough exceptions that I don't view one's degree-holding status as necessarily particularly indicative of one's ability to meaningfully analyze a topic either way. One should always evaluate an argument on its own logical merits, rather than the source (indeed, to do otherwise is an appeal to authority, a long established and well known logical fallacy.)
  • Floating currency... a certain segment of today's intellectual society deeply dislikes the current system of fiat currency, disparaging it as 'fake money' created by nothing more than debt. While I'll be the first to admit that it's not optimal - continued economic growth requires a continually growing amount of debt to finance it -- it's still far better than the gold standard, or any other fixed value system. The reason is actually pretty simple. The amount of gold (or any other commodity resource -- oil, silver, whatever) in the world grows extremely slowly. New production is brought online; old production peters out; some gold is consumed as jewelry (or more recently, used in electronics), and the average net gain is usually only a tiny year-on-year increase in the amount of gold available for use as specie. This works fine in a medieval agriculture-driven economy, where the yearly gains in extant value are also very tiny, because they're inherently limited by the speed of human and animal labor due to lack of technology. But the annual growth rate in extant value in an industrial economy is far faster than the growth in the amount of circulating gold. Technology lets people be amazingly more productive, and an industrial or postindustrial economy simply creates far too much value far too fast for gold to be viable as a currency.

    This was, in fact, tried for many years. The net result was massive deflation in the 1800s and early 1900s. Think about it: Industrial technology produced more and more goods every year, but there was a fixed (or nearly fixed) supply of gold-backed currency chasing all those new goods. If the supply of something (the supply of goods in the economy) increases, and the demand (in this case, the amount of money that is legally allowed to pay for goods) doesn't increase, then each individual unit of supply is worth less per unit of demand. Deflation. Massive deflation. This was theroot cause of the Great Depression (though there were certainly many other contributing factors.) A medieval agrarian economy would produce new value roughly linearly, in line with a linear growth in the amount of mined gold; whereas an industrial economy produces new value exponentially, as ever more advanced technologies are built on top of existing ones. The money supply has to keep up, somehow.

    The solution was to increase the amount of circulating money far beyond what any commodity can back. Fiat currency. This is the only way that we know of to ensure that the amount of money can match the amount of value.

  • Paul Legato
    www.economaton.com

no changes in population, no changes in fashion, no changes in technology, no changes in medical care, no changes in childbirth rates - no changes of any kind.

You've just described the conservative utopia the wingnuts yearn for. In their utopia, the dollar would never change.

"It's how a pig-ignorant idiot would explain the problem."

Your attitude often gets in the way of the point you're trying to make.

My understanding was that the reason they went off the gold standard was that it was constraining the economy. There just wasn't enough money for the amount of trade that people wanted to do. So it seems to me that it might make sense to tie the currency to something such as population or a food like corn or rice, so that as the amount of labor and people wanting to trade increases, the money supply would increase.But then again I am a general crack pot (or atleast admit that theory to be so.)

First of all, the floating exchange rate has been in place for 30 odd years. Before that most currencies are defined to a fix quantity of gold and I guess problems started to arise after WWII when trade booms and arbitrage opportunities arise.

Another point worth considering is that commodity is basically another currency without interest. So defining any currency to a commodity is equivalent to pegging one currency to another. Whether it is defined or market driven it all boils down to what the consensus public think the currency is worth to them. It's like a dictionary trying to define all the words I guess. Either way there will be issues to deal with.

My $0.02 in CAD which is worth (?) USD$0.02

Mark W in Vancouver BC

doc,

You're right. While both vary relative to corn, the dollar varies more because the quantity of money in the economy is constantly expanding under the current monetary policies.

Mark,

Woodhill's article is pretty terrible, but you should lay off the accusations engineers being particuarly likely to spount nonsense outside their field, since you got a lot of things wrong yourself.

Why does currency fluctuate? Because the relative values of different valuables fluctuate

This can explain some fluxuation of prices, but the overwhelming majority of the fluxuation of nominal prices (particularly their steady upward growth) in the modern US economy is explained by inflation, which is the result of the expansion of the money supply and can occur independent of any shift in relative valuations. It's been said before and I'll say it again: Inflation is always and everywhere a monetary phenomenon. The fact that you managed to write a post this long on monetary policy without mentioning money supply effects is morbidly fascinating.

In response to this, America has started pumping lots of money into ethanol production. As a result, supplies of some agricultural products, like corn, have had their demand increase rather dramatically - and so their prices have risen. And that in turn is increasing the price of other things that rely on corn - like many foods. So if we pegged our currency to gold, we'd still have seen an effective change in the price of a huge swath of consumer goods, caused by the change in the value of oil relative to the value of gold.

This actually has relatively little to do with shifts in oil prices and a lot to do with public policy that mandates the usage of ethanol in certain contexts. The price of corn hasn't been the product of market forces in over half a century and is actively managed by the goverment.

Commodity-backed currencies are generally slightly deflationary since the supply of the commodity in reserves generally grows slower than the total value of the economy, so you have fewer units of currency chansing more goods.

The nominal price of food could increase under a gold standard, but it did increase because of inflationary monetary policy and would have increased a lot less under a commodity-backed currency.

Suppose we pegged it to oil instead - so a dollar was fixed as the price of one gallon of refined gasoline. The dollar would be worth significantly more now than it was 10 years ago - because oil is worth so much more. But because people are paid in dollars, that would effectively make US labor dramatically more expensive, which would make it very hard for us to export goods to other countries - the production cost of American goods would be so much more expensive relative to goods produced elsewhere.

Only to the extent that wages are sticky, which is in the short run (normally about a year for wages). Once the wages adjust, it doesn't matter, since the wages will change to reflect the real value of the currency. Gold is often proposed as a commodity backing because it rarely undergoes dramatic fluxuations due to uncertainty in pricing like oil does.

What happens then is that the dollar becomes an effective exchange medium: if oil becomes more scarce, then its value relative to other goods should increase. But the fixed exchange of the indexed dollar would mean that you could use it to "trade" corn for oil at a fixed rate. How do you make the dollar work when people are gaming it like that? You need to infuse a huge amount of capital into the system to cover what people are pulling out via the exchange system.

Not true. In order to create an arbitrage opportunity, you need to allow it to be exchanged for two different bundles of goods at the same time. If you set the bundle to the same bundle of goods used to measure inflation, you'll actually have zero inflation.

You're completely in left field about the capital vs money comments. Woodhill is closer to correct here, although his argument breaks down in a few places (I don't know why he's going on about capital specifically, for example). What he's pointing out here is that when you have a fiat currency, the real value will undergo fluxations related to the government expansion of the monetary supply that are large and don't impact commodity-backed currencies. In practice, the US monetary policy has been consistenly expansionary since WWII. While prices under a commodity backed currency do vary, it's much, much, much less so than under the current expansionary monetary system. As you see in this graph of inflation rates and this graph of CPI, almost every year after WWII the US has had substantial inflation. Compare this with the post 1820s (necessary because of the major change in banking policy that Jackson drove) 19th century, where inflation and deflation tend to balance out over the years towards a slighly deflationary net. I think you're misunderstanding Woodhall's stance, which just the classic gold standard position of securing relative stablility by assigning a commodity exchange rate explained via an incredibly crappy metaphor.

This inflationary policy does have costs. Specifically, since the kind of expansionary monetary policy that the government has engaged in increases uncertainty about the future value of dollars. This creates uncertaintly in contracting, since if you contract in dollars, you don't know how much it will be worth when the contract is paid off, so you don't know the value of the contract. This uncertainty is economically costly and people pay to eliminate it via hedging. While commodity backing doesn't guarantee a value, it does constrain it based on the quantity of the commodity that will be available for reserves, whereas changes in the valuation of fiat currency are theoretically limitless.

Not that there aren't different problems with commodity backing (it's vulnerable to shocks to commodity supply and liquidity traps for example). The difference between commodity backing and fiat money is actually less in practice than people make it out to be due to the issue of reserve requirements, which can be used to peform manipulations equivalent to what the Fed currently does. What the central bankers are trying to do (stimulate demand, decrease inflation, stablize prices, target an interest rate) matters a lot more than what kind of currency it's being done with.

-MattXIV, an engineer

Disclaimer: I'm an electrical engineer.

I don't know if engineers are more likely to claim knowledge they don't have -- I've heard a lot of people do that. But even though this guy's clearly wrong, I can understand a little bit where he's coming from. The impression I've gotten (someone please correct me if I'm wrong) is that macroeconomics is very political in nature, suffers from many ideological splits, and has problems with the evidence for many of its ideas. When I hear two people debating economics, they usually both claim the same pieces of evidence as support -- "A recession happened here, which was during President B's administration!" "But look, it was right after President A left office!"

Economic schools of thought are named after the places they're taught. What does that mean? If I go to two different schools will I get two contradictory sets of lessons? Maybe this perceived subjectiveness is just another artifact of bad media coverage or something similar, like with evolution vs. creationism. But with evolution, you can present some pretty hard evidence. I don't get the impression that economics works the same way (it is a social science and all). So it's tempting, coming from a field whose biggest (and only?) ideological split (Edison vs. Westinghouse) involved technical merit clearly and easily triumphing over politics, to say that macroeconomics is fundamentally broken and needs some good ol' fashioned engineering to straighten it out.

The mistake I think a lot of people make is thinking that they can logically reason out everything they need to know about a field from the outside. In reality, most fields are far more complex than they seem from the outside, and having a bit of crossover knowledge often seems to do more harm than good.

By Adam Haun (not verified) on 13 Feb 2008 #permalink

A slightly different way of viewing it that I prefer is to see a currency as simply another commodity. You buy it, sell it, store it, borrow it. But unlike other commodities it doesn't expire and has no use other than as an exchange medium. We could use the unit of pork bellies - or wheat bushels, or steel, or gold - just as well, except that the supply is arbitrary (depending on weather, or mining output), expires (in the case of farm products, and to some degree steel), and is frequently used for other uses than as an exchange medium by its buyers.

Switching to, say, gold, would still give you all the perceived problems with a floating currency, and add all the issues with using an actual, rather than abstract, commodity as the exchange medium as well.

"Just as when a company prints new shares (and gives them away), it dilutes the value of the old shares, so when a country prints new money (i.e., deflates the currency), it dilutes the value of the old money (inflation)."

That can happen when money is printed but the amount of wealth stays the same. However, in modern economies, money is added for investment, which creates new wealth. It is all about keeping a good balance.

It's time I put my music degree to work and started writing about aerospace design.

"It's how a pig-ignorant idiot would explain the problem."
Your attitude often gets in the way of the point you're trying to make.
Posted by: Anon | February 13, 2008 6:15 PM

No, the point is perfectly clear, but some people seem to think pointing out someone who has made a pig-ignorant idiotic statement is a pig-ignorant idiot somehow detracts from the point (frequently expressed as the poster being "rude"). Some of us think it's important to point out when idiots are saying idiotic things. If it happened with more frequency, it might reduce the adoption of idiotic ideas.

What about the European experiences of forming a common currency? It seems relevant to the arguments here.

A currency fixes values for a community. If two communities decide to lock their currency together then the currency is averaged over a larger community. In Europe, the requirement for joining the Euro was/is that the economies that merged met some stablility criteria and had a fairly low inflation rate. The political goal was to build a trading block that could rival the USA with a common currency, alignment of taxes and tariffs and low internal market friction.

As thought experiments, consider how a community as large as the USA manages to have a common currency in the first place. For example the price of houses or gas in Texas is a lot less than in California. The geographic friction in the market supports this difference. However the price of things that can be easily ordered over the internet and shipped are essentially constant on a nationwide basis.

The question then is around the benefits of scale, if bigger is better then it would ultimately be good to lock the Euro and the Dollar into a single fixed market. If smaller is better then it would be good for California to create its own currency, so that it can float relative to Texas. Apart from history and politics (which really dominate what happens) is there an optimal size for a currency?

As friction is taken out of the market, by reduced transport and shipping costs, the Internet and the growth in non-physical goods, I expect that the optimal size of a currency increases.

So maybe the original statement is not so pig-ignorant idiotic after all?

By AdrianCockcroft (not verified) on 13 Feb 2008 #permalink

I have a degree in underwater basket weaving, and I'm also pignorant, but guess what! This gives me the right to complain about anything I damn well please, even if it has nothing to do with the hydrodynamics of the loom.

Einstein had it All Wrong, clearly. If you were moving at the Speed of Light, and you threw a Ball forward at 10 Miles Per Hour, then the ball would be Traveling at the Speed of Light plus 10 miles Per hour. It's so obvious, ANY Idiot could have come up with it. Even the Skeptical Genius Kent Hovind agrees with me on this. Not only this, but you can't BEND Light. It's like trying to bend Water, or a Gas. Light isn't even like those, though. It's just Energy. Numerous physicists have agreed with my point on this; even Einstein himSELF admitted that Light was Energy!

And Obviously, evolution could not have occurred because The Bible Says So!

P.S. That entire post was meant as a joke. I went back and added random capitalizations, but I thought that might make it seem like a "serious" post from a real YEC, or Young Engineering Creationist. (No offense to other, real engineers)

What surprised me about the original post is that he doesn't use math like he clearly should - make up some stupid equation in an attempt to baffle his audience. "Oh yeah, MATH is involved, so it must be true, right?" "This guy clearly knows more than I do about the situation, if he can give us equations that model it. Let's listen to him."

Dang. I forgot to do that too. Oh well.

Mark,

Please let my comment out of moderation. I'm guessing it got held because it had multiple links, but I posted it last evening around 7 and it's still not out yet, which is annoying since it contained a response to a specific commenter as well as several substantiative criticisms of your post that took a bit of time to prepare. Now it doesn't seem likely to get posted until the comment thread is dead.

Re deflation:

Tell me about it. I work in an idustry that is essentially defined by deflation; disk drives. Every year we are expected to sell a drive with twice the capacity for the same price (approximately). Likewise with the computer industry as a whole, twice as many transistors on a chip for the same price every 18 months. And this has been going on for 40 years and the industry continues to grow. Hmmm, what's wrong with deflation again?

Reductions in price and improvements in capability is not deflation. Deflation is a decrease in the general price level, typically caused by economic collapse and lack of demand. Debtors are ruined as the value of liabilites increases and asset values plummet. The Great Depression was a Deflation - it was deflation that destroyed the banking system during the period, wiping out people's savings.

I'm not really up on economics, so let me see if I've got this right.

Deflation is bad for the economy because things which are tied to a fixed number (like debts) essentially become worth *more*, while things that must have their value negotiated (anything you want to sell) lower their price as dollars become larger.

Inflation is bad for the economy in reverse - assets sell for more, but the amount of money you have to *buy* assets with (pay-contracts and savings) remains fixed in number, so that its value decreases as the dollar shrinks.

I'm surprised that Woodhill didn't realize his error immediately upon thinking up some of the examples that he did. One in particular, about the garage no longer fitting on the lot, sums up his error extremely well. For this to happen, it would have to be true that there is no *objective* measure of something's length - instead, the length of things is negotiated individually.

It's clear that we don't negotiate the length of items, but we *do* negotiate the worth of items. There is no objective scale measuring value - if there was, we'd just use it and have the stable utopia that Woodhill is thinking of. Instead, things change in value (as Mark said), and they do so independently. Items age and become worth less, while other things are built from cheap base materials and become worth more. That makes a fixed currency completely untenable.

One could argue that all we need are regular normalization events that reevaluate the worth of goods and assign them a price based on what percentage of the world's worth they contain, but as noted before, there is no objective scale to measure worth on.

By Xanthir, FCD (not verified) on 14 Feb 2008 #permalink

I'm surprised that Woodhill didn't realize his error immediately upon thinking up some of the examples that he did. One in particular, about the garage no longer fitting on the lot, sums up his error extremely well. For this to happen, it would have to be true that there is no *objective* measure of something's length - instead, the length of things is negotiated individually.

The issue isn't that there isn't an objective measure. The real values of a good can be measured objectively; nominal exchange values (what a good is bought and sold at) are objectively observable and can be combined with an inflation rate determined from a price index. There's no objectively correct choice of items for a price index, so there's no one true inflation rate, but any one that's broadly representative of goods that are sold in the economy will give you an approximately correct value that can be constructed based on objective observations. Thus, it is possible to meaningfully say whether the real value of a good has changed. Even changes in the real values of the same nominal quantity of currency can be observed via this method. A common unit of value needs to be adopted to make comparisions (normally the real value of a dollar in a certain year), but that's no different than having to compare feet to feet instead of meters, since it's possible to convert to any other good.

What Woodhill seems to be getting at here (the choice of metaphor is poor) is that relations fixed relations between real values and nominal values (for example, a contract to sell someone a certain bundle of goods for a certain price) can change without a change in the real value and that the current monetary policy calls for a steady drumbeat of inflation, ensuring significant downward changes in the real value of a specific nominal amount of currency, and that large net changes of this kind are difficult to predict or plan around. I don't blame people for not picking up on it, because unless you already are familiar with the argument, you're unlikely to recongize the house being a metaphor for it.

The main problem with the house metaphor is that real values (the size of the house) can change. Mark noticed this but failed to realize that while it's fatal to the house being a good metaphor, it's not fatal to the argument that chainging the the real length of the yard is a bad thing; indeed, it's not even germaine to it.

On the topic of deflation vs inflation, the annualized inflation rate for 1830-1900 is -0.13%. Between 1968, when the gold pegging of the dollar under Bretton Woods (a semi-gold standard system where most participating currencies were non-exchangeable but adjusted to keep their exchange rates with the $ and gold in a certain band) broke down, and 2000, the annualized inflation rate was 5.12%.

Thus, it can confidently be said that commodity backed currencies have much more stable real values. As for how bad inflation is relative to other monetary issues, economists disagree, and we'll leave at that.

(stats from here)

The issue isn't that there isn't an objective measure. The real values of a good can be measured objectively; nominal exchange values (what a good is bought and sold at) are objectively observable and can be combined with an inflation rate determined from a price index.

You misunderstand what I was saying. The fact that you can ask people what they think something is worth (or use a proxy for their opinion, such as what they actually pay) doesn't mean that there is an objective value. If you ask different people in different situations, the value will change.

On the other hand, length is truly objective. Length is the same no matter who measures it or when they do it (modulo relativistic corrections, obviously, but these are also well-defined).

As noted previously, identical land and houses in Texas and California can command vastly different prices. Value is a subjective quality.

By Xanthir, FCD (not verified) on 14 Feb 2008 #permalink

Woodhill may have made a pig-ignorant idiotic arguement, but I'm not so sure that our current system is any less pig-ignorant idiotic.

My understanding of the basic arguement here is that Woodhill wants to base money on a commodity. Mark points out that money based on a commodity is an awful (pig-ignorant idiotic) solution. However, credit and fiat money has its own problems. The Great Depression was based on abuse of Credit Markets. My fear is that our monetary system faces the same risk of a catostrophic failure.

Is there a way to base the value of money on per capita GDP?

David wrote
Is there a way to base the value of money on per capita GDP?

Milton Friedman has maintained that a mechanical growth in money supply at 3% or so - equal to GDP would be an ideal system. (you would want total GDP not per capita - if the population doubled with no change in GDP you would want the money supply to remain constant)

On the other hand, length is truly objective. Length is the same no matter who measures it or when they do it (modulo relativistic corrections, obviously, but these are also well-defined).

Doesn't this lead nicely into discussion of mathmatical measure theory?

By Baldeagle (not verified) on 15 Feb 2008 #permalink

I for one think the harsh criticism and 'idiot' remark were well deserved and would be unhappy if you stopped calling things like they are.

Having said this I have two quick comments.

First, while one can't fix the purchasing power of the dollar in more than one good simultaneously there is no block (in theory) to eliminating fluctuating exchange rates between currencies. Of course simply trying to peg your currency to another currency via your governments unilateral fiat can fail (when trust in your governments ability to fund conversion fails) if governments were willing to work together you could create a single worldwide currency and merely print it in different national styles with different relative values, i.e., the way they did with the euro before they eliminated the local physical currency.

Secondly I think you made a slight error in your example about setting the dollar to be the combined value of a basket of goods. If it is really the *combined* value then it doesn't fix a particular rate of exchange between oil and corn but it doesn't fix the amount of oil/corn/whatever the dollar can but either. For instance if the price of corn increases than that 1/2 a bushel of corn would become a larger fraction of the total value of the dollar and allow you to purchase more oil with it.

I was extremely proud when my son, at age thirteen (when he was already in university) gave the correct answer to the classical Economics question: "What is a dollar worth?"

Answer left to the student; I'll jump back in if nobody gives the correct answer this weekend.

Now that he's in his 2nd semester of USC Law School (aged 19) he's well on his way to having more dollars than his two professor parents combined. At least that's what the financial aid/loan folks are counting on.

This is an expansion of the Salem hypothesis.

By Unsympathetic reader (not verified) on 16 Feb 2008 #permalink

Mark, I think you are harsh in your comment. The article which you criticize has a point, although it may not formulate it in the best way.

The whole question starts with "what is money?".
It's actually a confusing and difficult question.
First, money today is a defined differently than a century ago.

A century ago, money and capital were strictly equivalent, by definition, for some certain goods with good properties (not too rare, not too common, can be divided, ...), such as gold or silver.
At that time, different countries moneys would not fluctuate relative to each other, because in reality they were all fixed in terms of weight of gold or silver.

That's something that the said article gets right. If you define your money that way, then there is no fluctuation between countries, unless different countries choose different goods (gold for one, oil for another, silver for a third, etc.).

Money is not defined that way anymore. Actually, money today for a consumer (not a bank) is little more than paper, backed by nothing except law (it is legal tender, which you have to accept as payment).

The monetary system has evolved thru many different stages, as detailed in some of the pointers below. It is arguable that today's "floating currency" is better than last century's gold standard (except if you are a government looking for a quick shot in the arm).

Below are some good references to learn more about the subject. Both also indicate the moral hazards of letting money be controlled by government, and the economic dangers of having a central bank.
I must admit it's still a complicated matter for me, but I found both those pointers very useful.

What Has Government Done to Our Money? by Murray N. Rothbard, at http://www.mises.org/money.asp
also available in audio at http://www.mises.org/media.aspx?action=category&ID=92

Money As Debt - by Paul Grignon
http://video.google.com/videoplay?docid=-9050474362583451279

One more thing: I believe that the author's point in describing himself as an engineer is not to seduce his readers into uncritical thinking, but rather to compare how you can define money with how you define other units.

I have to say, I was pretty impressed by many of your posts and subscribed to your blog. This post however is pretty sad.

A little note: The foot is not pegged, it is relative! Even the relation of the foot and meter would appear different at high speeds because their endpoint would be at different locations in space. They cannot always be in the same frame of reference.

And I would tend to think that printing more money allows for more trade of capital, since bartering has gone wayside. I can imagine the trouble of buying a house if I had to give up my car, and all other belongings, for it.

You misunderstand what I was saying. The fact that you can ask people what they think something is worth (or use a proxy for their opinion, such as what they actually pay) doesn't mean that there is an objective value. If you ask different people in different situations, the value will change.

I think you misunderstand the purpose of economic analysis. It isn't to pronounce metaphysical truths about value. It's to describe how people behave. I'm interested in exchange value as an objectively observable characteristic that can be used to predict behavior. It could be called "faghr'agek" for all I care if it still had the same relationships with the behaviors I'm interested in. The exchange value of a good occurs where production costs and use value (aka supply and demand) meet and is an objective property of an economy that is very powerful for predicting behavior. What you're describing are preferences, which are a non-directly-observable property of an individual.

For fungible and liquid goods, and any good can be defined in such a way as it is fungable (ex if houses at X and houses at Y aren't perfect substitutes, you can treat houses at X and houses at Y as separate goods instead of treating houses as a single good), the exchange value alone allows the prediction of trading behavior, as the below examples demonstrate. Comparing houses in TX to houses in CA is literally like comparing apples to oranges as far as economics is concerned - while both houses/food, they're imperfect substitutes for each other.

Example 1: A nice black evening dress that exchanges at $250 is worth $0 to me in use value, since, pretty or not, I don't wear dresses. However, if someone offered to sell me one for $150, I'd still buy it, since I could re-sell it at $250. And if I came to possess the dress, I'd reject an offer of less than $250 despite its lack fo use value since it has a greater exchange value for me. Thus, even if my use value is less than the exchange value, I'll still value a fungible and liquid good up to just below its exchange value.

Example 2: A new standard Fender Jazz Bass exchanges at $400. I love playing bass and like the neck and tone of the Jazz bass, so even if Fender cranked up the price to $800, I'd probaby still want one. However, if someone offered me $500 for my new Jazz bass, I'd take them up on the offer, since I could simply buy another one at $400 and pocket the extra $100. Thus, even if my use value is greater than the exchange value, I'll still value a fungible and liquid good up to just above its exchange value.

When you introduce transaction costs, it gets more complicated, but that merely results in a fairly narrow band of exchange values. If the transaction cost is $5, for example, I would buy the dress for less than $295 and only sell the guitar for more than $405. In these cases, you can still potentially separate out the transaction cost from the exchange value.

In the above example, it wouldn't make a lick of difference in actual observed behavior if my use value for the Jazz bass was $1000 instead of $800.

Currency generally has negligible use value, especially reserve notes. Thus, the only basis of comparison of a good with currency is its exchage value. The weird thing about the supply of fiat money is that it isn't constrained by any intrinsic scarcity like other goods, since the quantity available isn't limited by the increasing amount of effort necessary to produce more units. Take away the supply side constraint, and you can move anywhere on the demand curve. In contrast, fixing the exchange value of a dollar to a certain amount of gold to make the supply of dollars scarce because of the need to hold gold reserves and a function of the economy rather than public policy.

Wouldn't having a single world-wide currency mean that it would stop fluctuating, since there would be no other currency to trade it for?

"Needing" to have gold reserves doesn't sound very ideal to me. Gold is spread far and wide and hard to accumulate to a reserve, not to mention that during 20th century gold has become a commodity that has many industrial uses. Locking it up wouldn't be good for industry.

This gold standard advocacy to me sounds like all the gold speculators have dumped bunches of money into gold and now want to make it worth even more by getting government to buy it back at $10,000 an ounce, in order to maintain the amount of cash reserves in this new gold-based system.

It's strange - the fluctuating dollar is mainly about international exchange rates, yet Louis Woodhill is obsessed with a gold standard and Mark CC and the commentators are all ignoring international currencies and trade in their discussions.

From an engineering viewpoint, the problem is simple. The only way to determine an item's value is to put it on the market and determine its price. No sale, no price, no value. It's like quantum physics. There are things we can measure, and things we have to estimate based on conditional probabilities.

The reason we have currency is to allow people to store value in a convenient, relatively permanent form. (If we choose a cow standard, we have to take into account the fact that cows can age and die.) Gold is easy to assay, and it is stable, but it is not as simple as all that. There were multiplier effects. Even though gold sold for less than $1000 a pound, a pound of gold in reserve meant that a bank could issue $500,000 in loans. This was based on analysis of probable bank transactions, and was chosen to provide both a good chance of solvency and a good supply of currency so that the economy could function. (Anyone who understood banking in the 1930s could have come up with Feynman diagrams).

As with any engineering decision, circumstances can change and assumptions can be proven wrong. The newer Tacoma Narrows bridges don't use the original Galloping Gertie design.

Our current system is just as arbitrary, but we dropped the gold in the basement part. Look at the Federal Reserve devaluing the currency to cover the collapse in real estate prices. If you read the details it is something like "double secret probation".

MattXIV wrote, "On the topic of deflation vs inflation, the annualized inflation rate for 1830-1900 is -0.13%. Between 1968, when the gold pegging of the dollar under Bretton Woods (a semi-gold standard system where most participating currencies were non-exchangeable but adjusted to keep their exchange rates with the $ and gold in a certain band) broke down, and 2000, the annualized inflation rate was 5.12%."

And how does that compare to the inflation rate of the Deuschmark over the same period?

One problem with the Fed is that it's sole purpose is not to monitor the currancy supply. The Fed also has a mandate to 'stimulate the economy'.

In an ideal situation there would be just enough currancy added to an economy to cover every desired transaction. Too much additional currancy devalues it, leading to inflation, too little currancy results in deflation - every dollar is worth more.

The Bundesbank managed to avoid inflation in Germany after the Bretton Woods system broke down by keeping a careful eye on the money supply and adjusting their lending rate.

At various times, USA administrations have requested the Fed to loosen the money supply by lowering interest rates. Lower interest rates mean more borrowing against the reserves of a bank (to the limit allowed by the Fed), and automatically generates more currancy (without the Fed having to print a note).

This additional currancy will very likely be greater than the expansion of the GDP and inflation occurs, driving up the prices of goods. In an ideal market, and using a good with a fixed cost of production as a referance point (like gold) the result is that the prices rise at the same rate as the money supply is in excess of GDP growth. Practically, there are delays in how rapidly an economy corrects to fluctations in the money supply.

Which suggests that long term economic stimulus cannot be created by adjusting the lending rate or the money supply. A short term growth in business will be eventually offset by inflation, with the additional worry of business failures which were encouraged to invest in riskier businesses due to low lending rates.

Needless to say, this is true whether a currancy is pegged to a commodity or a fiat currancy. The mechanisms are slightly different for changing the level of currancy in an economy, but the results are the same.

I disagree with Milton Friedman a little in that as I see it, a 3% yearly increase in the supply of currancy will be far greater than GDP growth once you recognize than a unit of currancy may undergo a number of transactions every year. I suggest that there is a statistically normal distribution in the number of yearly transactions of each unit of currancy and your injection of money should account for the mean number of transactions.

And this all develops from looking only at currancy liquidity inside a single economy. Add in the other factors like the value of goods, and how that changes due to demand, or the relative value of competing currancies on the world market, and the complexity of the problem increases quite a bit.

Just my $0.02, which should probably be discounted by the fact that I'm an engineer. ;)

"A horse! a horse! my kingdom for a horse!"
(Richard III, by William Shakespere)

... I think that quotation demonstrates the fluctuating value of various items against each other according to supply/demand very well.

What you criticize in your blog post is not what the "engineer" author was saying, so I think you were overly harsh. I don't think he was arguing that pulling currency off the market by tying it to another commodity would stop fluctuation in currency value. He seemed more to be arguing that pulling currency off the market would stop the inflation/recession oscillation that is caused by government caused oversupplies and undersupplies of money (tied to the Federal funds rate).

Woodhill believes that the Fed is not very good at controlling the inflation rate, and that instead it tends to exarcerbate oscillations because of unintended consequences. He thinks maintaining currency value would better be left to the market itself, which is what tying the currency to something in the economy would do.

Is he right? I don't know, but his analysis is not as simplistic as you seem to think it is. And it is very free-market in philosophy.

vssg72

One thing that many people here seem to have overlooked is the vast increase in the supply of gold in the nineteenth century. California, Australia, the Klondike and South Africa were the main sources of this increased gold supply.
I have no academic qualifications but I wonder if this increase in 'money supply' helped sustain the Industrial Revolution.