Last week (2011 Feb 06-12) the question of backing the US dollar with something of value rose up yet again. Last Tuesday, Paul Ryan (Rep. WI) asked (or rather lectured) Ben Bernanke why the price of gold was rising, or conversely, the value of the dollar was dropping. He stated that some would see this “as a vote of no confidence against fiat currencies.” Fiat currencies are those that float, that is, are evaluated in reference to each other, not to any invariant commodity, like gold. This was such an event that by week’s end (Friday), the New York Sun had an editorial on it, and Paul Krugman featured him in his excellent column in the New York Times. Gold-denominated currency is again in the news. Not for the last time in this congressional session, I fear. My thinking (below) is that metals denominated currency is a bad idea and not the way to go.
Silver and gold both have been used to support the value of currency, on and off, at various times. In the 1700’s, the gold to silver valuation was about 15:1, reflecting the centuries long nearly constant price ratio of these metals, and almost the natural abundance ratio (1:17) of the two metals . This led to the metal standard: Sometimes gold denominated the currency, sometimes silver, sometimes both at the same time (bimetallic standard). Republican Abraham Lincoln (1862) was the first president to issue paper currency without any metal backing (the “green back,” to pay war debts). Democrat Franklin D. Roosevelt (1934) broke the link between a gold denominated dollar and domestic currency (but silver backed the dollar at home and international debts were still backed by gold). Democrat Lyndon Johnson (1964) ended issuance of Silver Certificate dollars and the last day one could convert a dollar to silver bullion was 1968 June 24. Republican Richard Nixon (1971) finally ended any and all connection between the US dollar and any metal of any sort.
The history of monetary standards is a tangle of the various political parties, and various monetary support strategies (gold, silver, other, and even none: where the value of the dollar is worth exactly one dollar).
The heart of the matter. A math model is developed in the Appendix (this posting) that shows how to relate inflation in a country to its currency. Only if you can denominate the currency by a commodity that does not change in intrinsic value over time, can inflation and currency evaluations may be done with confidence. If the basis for the denomination is not invariant, no accurate relation can be made determined. Our task – identify an invariant item to base our currency on. If this were possible, then our currency would be invariant, too.
If gold is to be our standard, then consider: it has risen from a 2001 starting point of PS(2001)=$300/oz to a current 2011 value of PS(2011)=$1370/oz in a classic exponential rise. This happened in the last 10 years with almost no record of inflation in other goods. I will return to this fascinating graph later, in a political posting. Without matching inflation, gold could not be even approximately invariant.
This last observation means gold is not a good candidate to denominate the dollar, since it is not steady enough. (Sorry, Rep. Ryan.) We can use something as a monetary basis only if its intrinsic value does not vary ever. If we can find something invariant, the math says that we should definitely back currency with it, whatever it might be.
The gold/silver ratio was nearly invariant for almost 300 years (1600-1900) and there was a lot of support for a bimetallic basing of all currency. Unfortunately, that was a unique historical epoch and the ratio wiggled about during the last century. look at these two graphs, set on the same time scales for comparison. These graphs are not corrected to 2011 dollars. If this had been done, the 1980 spikes would be much higher.
The oscillations of the two metals means that they cannot be used as a currency base … we must not denominate the dollar on anything this volatile! Those who would do so are setting the stage for monstrous scams that will rob us all, everyone but the Madoffs or Hunt Brothers in our population.
Proposals for an invariant: Count the many ways: If not (1) gold or silver, there must be some other invariant we could use … (2) The Coca-Cola standard: I remember being lifted high so that I could put my nickel into the slot and get a wonderfully cold bottle of “coke.” Now, you can pay $1.35 for a bottle anywhere. At one time I used this as a defacto inflation measure. Volumes changed, corn syrup replaced expensive sugar, other changes happened. No, soft drinks are not a good inflation indicator.
(3) Price of a new car for a middle class family of 4? About $200 in 1910, maybe $30,000 today. But now, the car is not manufactured in the US, it is just assembled here (read… insert part A into part B, weld to part C…). Much has changed and the cost base is controlled outside of our own borders. Too, the goodies expected for a standard car are much different… We cannot use this vehicles a reasonable inflation gauge. (4) The median week’s wage used to buy a pair of shoes. True for centuries. 40 years ago shoe repair shops were everywhere, it was cheaper to repair than to replace. Not now, for many shoes. Scratch median weekly wage. (5) Cost of a party dress? 40 years ago, most were made in the US and pricing was part of the local financial feedback loop that let us tie inflation to money. (6) How about a standard meal? (7) Or the price of ASTM 1010 steel? (8) Is a man’s dress shirt actually invariant? (9) What about a middle class income? This actually begs the question (means: forces the test) on what is meant by “middle class.” A middle class wage earner in the 1950’s could support a family of four on 1 salary and send the children to a university. Now a family of 4 with with median income from two salaries cannot send their children to a university. So – don’t go denominating your currency on a middle class family in the population.
We manufacture almost nothing nowadays. This breaks the loop that lets us use most fungible items as standards. Low prices, currently the norm in the US, are not due to local inflation and the value of one USD. Things that control costs are mainly the nearly slave-labor wages paid in Southeast Asia and China. Oh yes, and also are the working environments that have no expensive worker safety enforcements such as our national OSHA, EPA and other regulator/enforcement agencies. Our (10) cell phones and computers are cheap because China makes them and supplies most of the World’s volume of rare earth minerals. Workers must live in the toxic mining environment; short painful life = cheap chips for the rest of us. Makes good business sense, maybe, but rare earth minerals would be a lousy commodity to denominate the US dollar.
Ten failed proposals. Can you identify an invariant quantity that, if used, would allow currency, itself, to have a constant value?
Reconsider: What are we trying to do here? Denominating currency by an invariant quantity would let us evaluate inflation/currency relationships. Instead, if we wanted to compare different currencies, it would be much better to float all currencies. This lets the free and unobstructed marketplace do the comparisons. (How could Rep. Ryan not like that?) Restate: to compare different currencies, they should all float against each other. Invariant backing is only useful to understand inflation.
Two invariants are needed. The scientific/engineering truth is, we actually must use at least two commodities that never change, so that we can calibrate (verify) one against the other. At last! We would be able to evaluate currency without regard for local inflation.
We really do need the 2 invariants. If we tie currency to only one we will lose control: A single invariant denominator for the dollar would not allow consistency checking; it would allow circular reasoning that could cause rocketing highs and massive depressions. The charts show that gold and silver have a malleable value and are no longer constants; they make lousy dollar metrics.
This is unfortunate. I cannot identify even one invariant quantity, let alone two.
Denominating currency with fungible goods will lead to the centralization of wealth and opportunity, ultimately the return to global feudal societies. This is because market players with massively high wealth could (therefore would) manipulate the denominating item and end up controlling great portions of the currency pool.
Every post needs a bottom line. Here is mine: Smooth sentences with pretty words are easy to bend and twist into reasonable sounding traps. My own analysis (this post), says that gold denominated dollars not not any kind of way to go. This would, however, be good for speculators, or for those planning to corner the metals market. In other words, it is a good way for a fraction of a percent of our society to make their family fortune, but it would speed the ruin of our society. Krugman clearly wins on the gold standard issue.
Reality is its own thing. Reality operates as it will, not as ideologues order.
Charles J. Armentrout, Ann Arbor
2011 Feb 11, Minor additions 2011 Jun 03
Listed under Economics
Have a comment? Click on the title of this post, go to bottom, let us know.
Related posts: Click the INDEX button under the Banner picture
To refer to value, you must use some kind of standard along with arithmetic or algebra.
When you study inflation in school, you read that the price for an item will rise over time. If Country A experiences 10% annual inflation (0.1 rise in price per year) then something costing 10 markers this year will cost 11 next. This continuous process is easily measured, and unquestionably occurs.
Inflation is another of the exponential processes that I have been discussing (see Chess Fable and Rabbit Growth). It is clearly exponential because the amount by which price of an item increases next year depends on its current price this year. It is an example of unconstrained growth and clearly cannot continue forever; any extended period will lead to the destruction of the country, one way or another.
The math class: You are sometimes asked: if there was measurable inflation, how did the currency decrease in value? The instructor means – if price P in dollars/item rises as P=P0e rt (standard economic inflation formula for prices) then it should be obvious that the currency’s value falls with the negative rate of inflation: V=V0e–rt where V is the value of the monetary unit (items per dollar), and V0 is its value when the clock started (t=0). If the price starts to climb, it must be because the currency is dropping in value – you might have to define something like a “typical average item” to place a value on money, still, this argument probably feels right. But, is it so?
People get trapped in verbal embellishments. The speaker sets up both the situation and the discussion so that one’s sensitivity is numbed. Amazing hogwash can verbally pass as truth. Math is the tool to start from somewhere and automatically finishes up at the valid end point – or one as valid as the original assumptions. Math logic is a bright spotlight capable of shinning into murky corners.
Check the Inflation Calculator that gives inflated prices or value of currency, from the US Bureau of Labor Statistics
Let’s assume that there exists an item (or commodity, or material … ) whose value is constant for all time, unchanging in the tides of history. Use G to refer to the evaluation of this invariant kind of goods. N will stand for the number (amount of gold, number of wheat grains) of G item being used.
Eqn 1: N·G = P·V
The right side P·V is Price changes [dollar/item] times the currency value [item/dollar] so the product has no unit. The left side of Eqn 1 is the same as the right side (this is what “=” means). Since N is counting number like 17, the goods G must also be a pure number. G represents the pure, eternally unchanging standard value; N lets you specify how much of the invariant standard is being used to evaluate the currency. The standard horse would be 1 horse. Gold exists as atoms; perhaps the standard should refer to 1 atom, 1 troy ounce, or 1 metric ton of gold. What ever, you need the evaluation G times the amount N to balance the right side that has price times currency value (PV).
Implications of an invariant: An invariant is an item whose price rises to balance the fall in monetary value. Call this standard S. At some point in time (call it t=0) the Price for the standard is, for example, (PS)0= ($110)/(each unit of S) , the dollar valuation is (VS)0 = (one unit of S)/($110). Eqn 1 would say , N·G = (PS)0 (VS)0 = ($110/1 unit) (1 unit/$110) = 1. (use Eqn 1.) Because the denominating standard is invariant, this is its value forever. N=1 because we had “1 unit” as the amount of S to use, therefore G=1 and is unit-less.
Suppose at any other time t, we find the price of S rises above that $110 value at a constant annual rate: (PS) = $110e rt. Eqn 1 would have 1 = $110e rt· (VS) and VS = (1/110)e –rt . Thus our statement that you need an invariant commodity for the dollar base so that there is a clear understanding of what happens to monetary valuation when inflation happens.
For any other item, N·G = P0·V0 at a starting time t=0. Because the dollar is denominated by a true invariant, 1 = P0·V0 .
Summarize: Eqn 1 becomes
Eqn 2: P0V0 = 1 = P·V
This says that at future times, price inflation varies inversely with the monetary valuation. Quibble: in real life, this would not be true for each and every item. You would need to get a many-item average to apply Eqn 2.
Comment: Without that invariant standard S, we cannot logically tie inflation to the value of money. This is a strong statement. Eqn 1 would be true, of course, but it would be different for every item– some would rise in price, meaning the dollar fell for that item. Others would fall in price, meaning that the same dollar rose in value. It would not make sense to speak of an overall worth of the dollar, unless you pegged “worth” to relative worth compared to other currencies buying the identical quantity of the same items.