Why Citi Bank Chief Economist Willem Buiter’s Attack On The Swiss Gold Vote Is Gibberish

Normally this time of year I would be out on my land looking to ‘harvest’ some venison for the winter but the 5′ of snow outside my cottage has gotten the better of me today.  Instead I had a read through Zerohedge this morning and found an article  referencing Citi Bank Chief Economist Willem Buiter.  The article basically points out that Buiter is making a fool of himself and the article is quite convincing.  But I want to dig into it a little deeper because I honestly believe this Swiss gold referendum is one of those incredibly important moments in history that have the ability to change the world.  And the fact that guys like Citi Bank’s Chief Economist are pleading with Swiss to vote “No” to their upcoming referendum tells me I’m not the only one that thinks so.   And so I felt compelled to have a look at his arguments and what I found is a bit shocking.

Now Buiter is arguing against the value of gold and thus the value of a central bank holding gold.  Because he is directing his commentary at the Swiss ahead of their referendum he structure’s his arguments to coincide with the 3 stated aspects of the Swiss vote on gold to take place on November 30.  The three aspects of the referendum are as follows.

(1) the Swiss National Bank (the SNB) must hold 20% of its assets as gold, (2) the SNB has to repatriate the 30% of its official gold stock that is now held abroad by the Bank of England and Bank of Canada and has to physically hold all its gold in Switzerland, and (3) the SNB may never sell any gold again.

Buiter provides the following respective conclusions to each of the above aspects and then goes on to explain each conclusion in some detail.

(1) Gold has no intrinsic value but is in a 6,000 year old bubble and it is unsound for a central bank to hold 20% of its assets in any single commodity even if it has some intrinsic value. (2) It is unsound to forego the custodial risk diversification of holding gold in various countries. (3) If the gold holdings can never be sold then it is, in effect, worthless.

Now very quickly we can invalidate most of Buiter’s rebuttals.  Let’s jump to the second and third aspects of the referendum.  Buiter’s rebuttal to second aspect of the referendum to not repatriate gold remaining with foreign nations is premised on custodial risk reduction.  Having lived through the Refco and MF Global implosions I am very conscious of custodial risk.  However, it would seem that holding gold with a foreign bank also carries significant risk.  For instance, when Germany demanded its gold back the US was either unable or unwilling to do so.  And so if there is risk either way, I would choose to be in control of my gold as I have more incentive not to lose it than does a foreign bank to whom it doesn’t belong.

Buiter’s rebuttal to the third aspect of the referendum is that if gold can never be sold they are making it worthless.  Well the referendum is making it so the Swiss National Bank cannot sell the Swiss people’s gold.  The Swiss people will still have a process by which they can sell their gold holdings if it is deemed in their best interest.  And so that is a ridiculous argument.

Let’s now take the second part of Buiter’s rebuttal to the first aspect of the Swiss gold referendum which is that central banks should not hold 20% of their assets in any single commodity.  He really doesn’t support this argument at all.  In fact, he seems to argue that gold and currency are both fiat which he seems to define as something with no intrinsic value.  But if gold and currency are essentially the same asset he doesn’t explain why it would be ok to hold currency but not gold.  Also the Swiss already hold more than 7% of assets in gold and so if it is a problem with the allocation amount, he really does nothing to differentiate between the 7% and 20% so really there isn’t anything in that to refute as he doesn’t give a reason.

But quite ironically what Buiter does do in his attempt to prove that gold is really just a fiat commodity currency no different from any fiat currency is to prove the exact opposite and thus provides the Swiss with the very argument they need to vote “Yes”.

And so let’s have a detailed look at his proof that gold is just another worthless fiat currency.

Now Buiter admits that as long as there is risk of inflation by traditional fiat currencies gold is an attractive store of value.  However, he suggests the assumption that gold can hold its value while fiat currencies lose their value is an unwarranted assumption.  He attempts to prove this by taking us through a parallel world.  In his model world we have a static economy.  That is, Population, endowments, technology, government spending, taxes and preferences  and fiat money (which could be gold) are all constant. The government budget is balanced and prices are flexible.  Speculation does occur and thus speculative bubbles known as non-stationary equilibria are present in this model economy.

Buiter suggests that this model economy will have positive constant general price level which economists call fundamental equilibrium.  He suggests that regardless of the initial general price level, that given time, because of speculative bubbles the general price level will rise.  And because holders of money will anticipate this inflation they will reduce the amount of money stock they hold.  Buiter suggests that over time there will ultimately be infinite speculative bubbles resulting in money holders reducing their money stocks to zero.  At that point the general price level becomes infinity.  He suggests that anytime the value of a fiat currency is above zero it is a bubble meaning that the natural fundamental equilibrium is zero because of the inflation expectation brought on by infinite speculation.  And so he is suggesting that because fiat currencies (including gold) have no intrinsic value their value is based only on perception.  As people expect infinite inflation due to infinite speculation their perception of the currency value drops to zero thus making its value zero.   His point is that gold too is capable of losing all of its value because it is intrinsically worthless outside of people’s perception.

The first problem with his argument is that 30% of gold is used in commercial production, mostly jewellery.  It also requires inputs for production and therefore it has a minimum intrinsic value of the average cost of production; by way of microeconomic policy it ceases to be produced below that point.  But I know there those out there who will still argue it really doesn’t have any good use.  And so for those folks I will walk through the second problem of Buiter’s argument while assuming gold has no intrinsic value.

The second problem, quite shockingly for someone in his role, is it seems that Buiter is misguided in his notion of inflation.  He rightfully suggests that speculation can lead to price increases but quite mistakenly suggests that price increases equate to inflation.  The two are certainly not the same.  You see inflation is a type of price increase.  Prices can increase two ways.  One is by way of supply and demand of the asset being purchased and the other is through supply and demand of the money being used to purchase the asset.  Prices will rise when demand for the asset grows larger (everything else staying constant) but that does not constitute inflation.  Prices can also rise when supply of money grows larger than demand for  money being used to transact the asset and that is inflation.  The problem for Buiter is that in his model economy the money stock (which he suggests could be essentially any fiat form) is constant .  Therefore prices can increase by way of speculation for underlying assets but inflation is impossible as money can be neither added or destroyed.

If we go further down this rabbit whole it actually leads us to the holy grail.  It proves how gold is different from a fiat currency, so let’s continue down the rabbit hole.  Think about the model economy he set up.  In order to maximize demand in such an economy you would have to divide the constant money stock equally amongst the population which is also constant.  This is because of the principle of diminishing marginal utility of consumption which says ceteris paribus, the wider the income distribution the higher the consumption.  Let’s say in this world we did have speculation.  A few people realized that demand for one asset would increase and so they bought lots of that asset.  It turns out they were right and prices of that asset moved up on the higher demand.  They then sold some of that asset at much higher prices than they paid.  Through this process they are accumulating a higher proportion of money stock than others.  And as that process is taking place, due to diminishing marginal utility to consume, total consumption (or demand for assets) in the economy is declining.  So while speculators put their newly earned excess cash in savings it reduces total consumption and the effective supply of money in the economy.  And so for the rest of the population, marginal utility for money is higher meaning the value of money has gone up while demand for assets has declined, this is what we call deflation.

And so what we see is Buiter actually proves that, even in the face of wild speculation, when the money stock is limited, deflation is possible while inflation is very unlikely.  This suggests that it is not the money holders’ perception of value, which could very well take place in the model economy, but is the supply of money that results in inflation.  It is this very notion that is the difference between gold and fiat currency.  For in this model economy because money supply was limited and could not be increased, money could not inflate to a zero value.  While in the real world because gold is finite and is costly to produce, its supply is limited and so, similar to the constant money stock in the model economy, gold cannot inflate to a zero value.  However, because fiat currency is infinite with essentially no cost to create, it can and will, according to Buiter, inflate to a zero value over time.  It is not perception of value but money supply and the predisposition of money supply to increase that results in devaluation.  And so there he has actually proven that gold is a better storage of value than fiat currency regardless of perceived instrinsic value.

But what about economic growth?  Don’t the central banks tell us that inflation is necessary for economic growth?  Won’t a lack of inflation then prevent economic growth?  Well let’s use Buiter’s own model to find out.  But let’s start back at the beginning of his model world again with all the same constraints.  Is it possible in Buiter’s world, where inflation is impossible, to have economic growth?  And if so then we know that the central bank theory of inflation as a necessary ingredient to economic growth is false.  What we find is that absolutely we can have economic growth in our model economy.  Economic growth is achieved through increased productivity which results in higher output without an increase in money supply.

Now in this model economy because money stock is constant, economic growth results in declining prices for assets by way of supply increases of (non-currency) assets.  You see the reality is that when money is produced rather than printed economic growth happens without inflation and in fact prices for goods can decline.  In the real world money supply can increase but when it occurs from real economic growth, price levels stay the same.  Think about what happened in the model economy.  Because of economic growth things got cheaper and people could then buy more things with the same supply of dollars.  In an economy where money supply can be increased by the amount of produced economic growth, the output is split between additional money supply (in the form of profits) and additional assets.  In that scenario prices still didn’t move up.  So when economic growth is achieved by production and not by printed money, prices stay the same or decline.  When money supply is increased by printing and leads to increased output, price level goes up and that’s called inflation not economic growth.

To bring this home for the Swiss people.  The takeaway from Buiter’s thought experiment is that because gold is limited and is costly to produce it has intrinsic value.  And because it has intrinsic value it is a much better storage of value than fiat currency.  In fact, Buiter proves that because fiat currency has no intrinsic value that ultimately it will inflate to a zero value over time.   And so Buiter attempted to convince you that a Yes vote for the referendum will leave you with a money stock that is intrinsically worthless, that you will be taking on extraordinary custodial risk and that you will be making your gold holdings worthless.  But what we’ve just done is methodically proven that the exact opposite is true.  That by voting Yes on the referendum you will have a money stock that is a far better asset to preserve value, that you will be reducing your custodial risk and that if you, the Swiss people, decide it is in your best interest to sell gold at some point in the future it will be there for you to sell.  I would conclude that when even the best and brightest economists in the world are at a loss to make a reasonable argument to vote no in the upcoming referendum, it becomes abundantly clear that a “Yes” vote is the right vote.  And I expect a Yes vote in the Swiss gold referendum does more to nudge humanity back on track than any other single event since the fall of the Berlin Wall.