How do you value currency?

[claphands22]

How do you value currency?

 

I read that the Canadian dollar is strong and the US dollar is weak and I understand why. The US government has a large amount of debt, large trade deficit, increased amount of dollars in the money supply....I get that. Yet, how do you intrinsically value currency?  Is it just whatever Mr. Market says it is? A_Basic_Bag_of_Groceries is worth 100USD and A_Basic_Bag_of_Groceries is worth 200XYZ. Therefore the intrinsic value of every USD is 2 XYZ?  No?

 

I feel like I am asking a dumb question, so if it is, just say so.

[ExpectedValue]

I think that valuing currencies is pretty tough... There are a couple books by Rosenberg, FX Determination and Currency Forecasting that deal with fundamental analysis of currencies. The fact is, you're going to be trying to figure out how two economies will be performing in the medium to long term, which is pretty difficult.

 

Most successful global macro guys don't only take long term positions but are rather in the markets day in day out, trading daily. What that allows them to do is constantly test their opinions so that they can change their positions if their hypothesis turns out to be incorrect.

[wabuffo]

Best way to value a currency is in relation to how many ounces of gold it will buy.  Gold isn't perfect but its own supply/demand characteristics are so stable that any change in gold's price is more a function of changes in the underlying value of the currency gold's price is quoted in than changes to the demand/supply of gold itself. 

 

A corollary to this point is that if you take currency out of the equation -- you can also get an insight into the "barter" price of other commodities in relation to gold.  For example, the relationship of a barrel of oil to an ounce of gold has been more stable (with some volatility) than the price of oil expressed in US dollars.  This helps one to understand how much of oil's rise in the last decade has been to a fall in the value of the US dollar than to any increasing scarcity in the supply of oil.  (hint -- mostly due to the fall of the US dollar).

 

Again -- measuring stuff by using gold as a "currency" isn't perfect and you have to look at longer-term trends rather than daily or monthly relationships.  Gold is as close to a North Star as you can get in this imperfect world, though even the North Star has a "wiggle" in its position in the sky.

 

A final point about taking the US dollar out of the equation.  Another interesting viewpoint is measuring the price of "stuff" not by the price measured in a particular currency, but rather than a unit of labor (eg, what an hour of labor will buy).  By this measure, you can get a sense of the growing standard of living over time that becomes obvious when you measure what an hour of labor buys these days vs 20, 50 years ago.  It'll be interesting to see if this trend continues or not in the future.  I sure hope it does.

 

http://www.dallasfed.org/fed/annual/1999p/ar97.pdf

 

wabuffo

[StubbleJumper]

A rough approach is to apply Purchasing Power Parity theory when valuing currencies.  These models would suggest that the Canadian dollar might be about 15% overvalued vis-a-vis the US dollar.....  For Canadians thinking about buying US blue chips like JNJ, PG, WMT, or PFE it might be somewhat reassuring that there is some prospect of currency reversion that might slightly juice the returns.  For those interested, this is a good site on PPP:

 

http://fx.sauder.ubc.ca/PPP.html

 

 

SJ

 

 

[WideMoat]

If currency is a claim on future labor, then a currency should be worth the value of that labor--adjusting for the size of the labor pool willing to work for it, the productivity of that labor pool, and the ease by which it is created (via credit, interest rates, printing presses, etc.).

[claphands22]

I think that valuing currencies is pretty tough... There are a couple books by Rosenberg, FX Determination and Currency Forecasting that deal with fundamental analysis of currencies. The fact is, you're going to be trying to figure out how two economies will be performing in the medium to long term, which is pretty difficult.

 

Most successful global macro guys don't only take long term positions but are rather in the markets day in day out, trading daily. What that allows them to do is constantly test their opinions so that they can change their positions if their hypothesis turns out to be incorrect.

 

TariqAli,

 

Yes, I absolutely agree. Currency seems be one of those complex-adaptive systems in which predicting the currency prices is extremely difficult.  Thanks for the book recommendation - hopefully I can pick up the book next time I'm in the states.  I'm really interested in an intelligent way to understand currency price. At the moment, I just take whatever Mr. Market feeds me. 

 

What hypothesis exactly would a global macro guy be testing, in which he goes in and out of currency trade in a day?

[claphands22]

Best way to value a currency is in relation to how many ounces of gold it will buy.   Gold isn't perfect but its own supply/demand characteristics are so stable that any change in gold's price is more a function of changes in the underlying value of the currency gold's price is quoted in than changes to the demand/supply of gold itself. 

 

A corollary to this point is that if you take currency out of the equation -- you can also get an insight into the "barter" price of other commodities in relation to gold.  For example, the relationship of a barrel of oil to an ounce of gold has been more stable (with some volatility) than the price of oil expressed in US dollars.   This helps one to understand how much of oil's rise in the last decade has been to a fall in the value of the US dollar than to any increasing scarcity in the supply of oil.  (hint -- mostly due to the fall of the US dollar).

 

Again -- measuring stuff by using gold as a "currency" isn't perfect and you have to look at longer-term trends rather than daily or monthly relationships.  Gold is as close to a North Star as you can get in this imperfect world, though even the North Star has a "wiggle" in its position in the sky.

 

A final point about taking the US dollar out of the equation.  Another interesting viewpoint is measuring the price of "stuff" not by the price measured in a particular currency, but rather than a unit of labor (eg, what an hour of labor will buy).  By this measure, you can get a sense of the growing standard of living over time that becomes obvious when you measure what an hour of labor buys these days vs 20, 50 years ago.  It'll be interesting to see if this trend continues or not in the future.  I sure hope it does.

 

http://www.dallasfed.org/fed/annual/1999p/ar97.pdf

 

wabuffo

 

Do you know any good books that explain the supply and demand characteristics of gold? I have no idea what those would be.

 

I really like your idea of using gold as a north star in terms of valuing currencies. I never thought about using gold as a test, in terms of, crude's rise in price. Very interesting.  Yet, if you are using gold as a valuation tool of the dollar - how do you know gold isn't overvalued?

 

I read the Time Well Spent article this morning. It was great article and now I believe Goods/time-work is a great metric for the rising standard of living. I enjoyed the part about education and medical costs currently outpacing inflation, and that was written in 1997! I wonder if education/medical care are another asset bubble waiting to burst.

[claphands22]

A rough approach is to apply Purchasing Power Parity theory when valuing currencies.  These models would suggest that the Canadian dollar might be about 15% overvalued vis-a-vis the US dollar.....  For Canadians thinking about buying US blue chips like JNJ, PG, WMT, or PFE it might be somewhat reassuring that there is some prospect of currency reversion that might slightly juice the returns.  For those interested, this is a good site on PPP:

 

http://fx.sauder.ubc.ca/PPP.html

 

 

SJ

 

 

 

SJ,

 

Using PPP as a tool to value currencies makes a lot of sense to me, in fact I like this idea lot. Yet, their valuations don't take into account the amount of currency held outside the country (which if sold would flood the market with dollars), debt/GDP or other countries destroying their own currencies.

 

Great link. Do you know if using PPP as valuation tool has worked in the past? Like if PPP was used a valuation tool 10 years ago, how well the results matched the valuation prediction?

[wabuffo]

how do you know gold isn't overvalued?

 

Because the total above-ground inventory of all gold ever mined is around 158,000 tonnes.  As you can see from this graph, total annual output from mining activities is 2000-2500 tons and is steadily rising (but no big spikes in new mine output).  As such the annual supply of new gold is only around 1-2% of total above-ground inventory. 

 

http://upload.wikimedia.org/wikipedia/commons/a/aa/Gold_-_world_production_trend.svg

 

Demand for gold is about the same.  That's why gold's value is so stable relative to any other commodities.  Its because gold's annual new supply/total inventory ratio is the lowest of almost any other metal or commodity.  Its always possible that in the short term, gold's price could also be volatile as could happen in any market where short-term prices reflect the market as a voting machine.

 

If gold supply is rising about 1-2% per year and world GDP is rising 2-3% per year, then gold's price should rise a bit every year due to its relatively scarcity in comparison to the size of the world economy.  So probably a bit of gold's price rise over the last 10-15 years is due to its growing scarcity -- and not -- currency debasement.  But in that case, gold should be at maybe $450-$500 USD per oz. 

 

FWIW -- its an imperfect measuring stick in an imperfect world, but gold is the best measuring stick we got.

 

wabuffo

[Packer16]

One aspect you need to take into account is interest rate parity.  Unlike PPP, for which the currency can deviate at times greatly, interest rate parity is know quantity which will determine the spot and futures prices.  The approach is to see if the results of the interest rate parity make sense with regards to PPP. 

 

Packer

[StubbleJumper]

A rough approach is to apply Purchasing Power Parity theory when valuing currencies.  These models would suggest that the Canadian dollar might be about 15% overvalued vis-a-vis the US dollar.....  For Canadians thinking about buying US blue chips like JNJ, PG, WMT, or PFE it might be somewhat reassuring that there is some prospect of currency reversion that might slightly juice the returns.  For those interested, this is a good site on PPP:

 

http://fx.sauder.ubc.ca/PPP.html

 

 

SJ

 

 

 

SJ,

 

Using PPP as a tool to value currencies makes a lot of sense to me, in fact I like this idea lot. Yet, their valuations don't take into account the amount of currency held outside the country (which if sold would flood the market with dollars), debt/GDP or other countries destroying their own currencies.

 

Great link. Do you know if using PPP as valuation tool has worked in the past? Like if PPP was used a valuation tool 10 years ago, how well the results matched the valuation prediction?

 

 

PPP can show undervaluation or overvaluation for long periods, potentially even a few years, but normally there is a reversion. As always, it would be of minimal use for short-term trading as the market can remain rational for longer than you can remain solvent.  However, if I had paid attention to PPP in 2003, I might not have aggressively bought US- denominated stocks if I had understand that I would likely be facing a 30-40% currency headwind. 

 

OTOH, looks like there might now be a bit of a tailwind for the next few years.

 

SJ

[shalab]

1000 tonnes of gold are held in gold etf reserves. Also, public sentiment affects gold demand/supply. If a lot of people are piling into gold, it is likely that there is already a bubble of some sort.

 

From wikipedia:

 

 

Given the huge quantity of stored gold, compared to the annual production, the price of gold is mainly affected by changes in sentiment, rather than changes in annual production.[11] According to the World Gold Council, annual mine production of gold over the last few years has been close to 2,500 tonnes.[12] About 2,000 tonnes goes into jewelry or industrial/dental production, and around 500 tonnes goes to retail investors and exchange traded gold funds.[12] This translates to an annual demand for gold to be 1000 tonnes in excess over mine production which has come from central bank sales and other disposal.[12]

 

[wabuffo]

Given the huge quantity of stored gold, compared to the annual production, the price of gold is mainly affected by changes in sentiment, rather than changes in annual production.

 

Well sentiment is just another way of saying "Mr. Market".  In other words, gold's price is a real-time market assessment of the value of currencies in relation to gold.  Market pricing, to be sure, can get out of whack occasionally, but I believe its generally efficient.  i.e, in the short run the market is a voting maching, in the long run its a weighing machine and all that....  I think you have to believe that like the stock market, Mr. Market occasionally gets it wrong but generally gets it roughly right.  Gold's key attribute is its absolute stability relative to almost any other commodity (whose inventory reflects consumption levels that keep the annual supply/above ground inventory ratios at much, much higher ratios than gold's ratios -- eg 1-2% for gold vs say petroleum which is often much higher than 100%).  Gold's stability is what enables its virtue as a store of value relative to currencies.

 

I do agree that gold's supply is probably not keeping up with world GDP -- so its price should rise slightly over time (perhaps 1% per year) relative to a theoretically perfectly stable currency to reflect its inventory growth being slightly less than long-term world economic growth.

 

Again - I'm not saying gold is perfect, but its better than comparing currencies to one another since they are all depreciating now -- just at different rates.  But very smart investors with good long-term track records like Paulson and Einhorn are buying gold these days so it depends how you feel about their "voting records" as a measure of their "sentiment".

 

I think its generally a problem for the world when money moves from productive capital investment to gold (which is about as unproductive an investment as one can imagine).  Its not healthy but its a signal that historically cannot be ignored.  For example, gold moved out of its historical range of $275-$400 per oz where it had stayed range-bound from 1982-2004 in late 2004, early 2005.  It quickly doubled at a time (2004-2007) when housing prices in the US boomed and CPI seemed to stay low.  And yet that "sentiment" pouring into gold was a foreshadowing of problems ahead. 

 

As it turned out, US dollar liquidity pouring out of the US domestic market into foreign hands (Middle East, China) combined with investment banks getting the SEC to loosen their leverage constraints (via the SEC's re-write of the 2004 net capital rules) created an environment where that liquidity poured into US house prices.  Gold's price rise from $350 to $800 during that same period was a huge move and showed that gold was indeed very sensitive to changes, not in its own supply/demand characteristics, but changes in the supply/demand of the US dollar in which gold's price was being quoted.  None of the traditional measures (CPI, Treasury market yields) signalled this -- but gold did.

 

After the turmoil of last fall and the epic battle between the Fed, Treasury and the emergent debt deflation during the summer of 2008 in which gold fell from $1000 at the time of the Bear Stearns debacle to the low $700s in November (signalling indeed a deflationary spiral was beginning) -- gold is now firmly on the upward march again.  That is signalling surging liquidity again -- where is that liquidity going to show up (besides in gold's price), I have no idea.  But my guess would be that it's not good news and that we are going to get reacquainted with the misery index again.

 

wabuffo