Friday, August 17, 2012

Interest rates and gold

Nick Rowe recently asked what the point of repoing an object was when you could just sell it, repurchasing it later. He could see three reasons. Firstly, you might repo a particluar thing because you hold it dear and want it back. It might not be available were you to simply try buying it back. Secondly, you might repo an asset because future liquidity might be an issue. Thirdly, you might repo an asset rather than sell it because future prices are uncertain. My comment used gold markets as an analogy:
Nick, I think for financial assets you are right about points 2 and 3. In gold markets, for instance, you'd rather lend or swap (ie. repo) your gold than sell it (upon the anticipation of buying it back at some future point) because you might fear that, come time to buy the gold back, the future price could be much higher, or that the gold market could be illiquid and you might not be able to buy.
Incidentally, you can also sell your gold and buy a futures contract. Selling spot and buying a futures contract is financially equivalent to swapping (repoing) gold - in both transactions you'll lock in a guaranteed price and will avoid the risk of illiquidity. So your question: why repo? is similar to the question: should I sell some asset and simultaneously buy a futures contract or sell it and take the risk of buying back at spot at some future point in time.
Some people might recognize these various gold transactions. A gold swap — a temporary exchange of gold for cash — is transacted at the GOFO (gold forward) rate. A gold loan, which is an unsecured and temporary exchange of gold for nothing but a promise to repay, is transacted at the gold lease rate.

You can actually conceptualize both transactions as swaps. In the first, gold is temporarily swapped for a liquid and safe financial promise (cash). In the second, gold is temporarily swapped for an illiquid and safe financial promise (a promise to repay the gold ie. "paper gold"). Lingo-wise a swap is no more than a repo.

The different rates at which these two swaps are conducted — GOFO and the lease rate — will be determined by the relative subjective value gold owners place on the swapped-for assets. Because liquid financial promises provide more services to their holders than illiquid promises, anyone swapping away gold will prefer the former to the latter. That's why the gold lease rate — the rate paid by the person taking the gold to the person foregoing that gold - has always been higher than GOFO. After all, the party swapping away gold needs some increased financial incentive to take the illiquid "paper gold" asset rather than the liquid cash.

Indeed, liquid and safe financial assets are so valued, and so easily storable, that in general, anyone swapping away their gold will not receive a fee for foregoing the metal. Rather, they will pay a fee for the advantages of getting cash. This fee is what GOFO represents.

These ideas can be tough to conceptualize. A while ago I caught the folks at FT Alphaville mixing them up. For instance, the author maintained that a dearth of cash in markets now meant that
since their cash had become more valuable to the market than their gold, they could now make a return from lending cash against gold, as opposed to gold against cash.
The above comment implies that the gold swap rate — cash for gold, or GOFO rate — has somehow switched. But in actuality, those swapping away cash for someone else's gold always earn a return (GOFO). This is because they are foregoing liquidity. The gold lease rate has switched, but that is a different rate, and a different story altogether.

On the topic of gold, David Glasner asks why gold markets are rising due to hyperinflation fears but bond markets aren't:
Now my question — and it’s primarily directed to all those believers in the efficient market hypothesis out there — is how does one explain the apparently inconsistent expectations underlying the bond markets and the gold markets. Should there not be a profitable trading strategy out there that would enable one to arbitrage the inconsistent expectations of the gold markets and the bond markets?
I don't think there is any discordance to explain:
I think gold prices and bond prices have been rising over the last three years for similar reasons. In general, the economy-wide expected rate of return has been falling (towards zero, perhaps below it) as investors grow fearful of the future. This pushes people into safe bonds. It also pushes them into assets like gold that have very low storage costs, since buying some easily-storable durable asset and holding it over time provides a 0% return, better than most risky alternatives which are expected to fall.
So I don’t think there are segmented expectations in these two markets, nor do I think it is worthwhile trying to arbitrage them.
Returning to the discussion of GOFO, if today's gold markets were actually dominated by those who believed in the inevitability of hyperinflation rather than (fairly) rational actors, then GOFO would be inverted... in essence, anyone who swapped away their gold for cash would expect to receive the GOFO rate rather than pay it. The current practice, as I pointed out above, is to pay that rate, not receive it. The reason for the inversion would be because the destiny of cash in a hyperinflation is demonetization and, as a result, it will lose all of its liquidity premium, whereas gold's destiny is to be remonetized and gain a relative liquidity premium. As a result, any gold-for-cash swap based on a universal expectation of hyperinflation would require whoever foregoes the benefits of gold's inevitable superior liquidity to be compensated by a return. Needless to say, GOFO has not inverted. Those foregoing  their gold still pay up to those foregoing cash.

Wednesday, August 15, 2012

Is the Swiss National Bank really Chuck Norris?

I once got accused by Scott Sumner of having the silliest comment he had ever read back on this post. Recent events show that I wasn't being so silly.

Around that time, the Swiss National Bank (SNB) had announced a peg of 1.20 EUR/CHF. The argument going around the market monetarist blogs back then was that central banks were akin to Chuck Norris - they only needed to explicitly announce a target and that target would be effortlessly hit, just like how Chuck Norris can make a row of kung-fu masters fall like dominoes just by threatening to hit them.

I made a few other comments to the effect that a central bank has to build up credibility before anyone will accept it as Chuck Norris-like. Here is one comment:
On August 3, the SNB announced it would be purchasing CHF50b in assets to drive EUR/CHF up. Over the next five days it purchased this amount, but the CHF continued to strengthen. On the 10th the SNB announced it would be purchasing an additional CHF40b in assets, which it proceeded to purchase over the next five days. The CHF finally started to weaken. On the 17th, the SNB announced it would purchase another CHF80b in assets. The data shows that it executed this full amount by the end of the month.
EUR/CHF went from 1.02 on August 10 to 1.19 by August 28, so a lot of speculators were hurt. It fell back to 1.10 in the next four trading days, and then on September 6 the peg was announced. Presumably the announcement of the peg drove EUR/CHF back up to 1.20 but seems not have required as much effort, although I'd wait to see the SNB's next monthly bulletin to be sure.
The point of I was trying to make back then is that what made the 1.20 peg credible was the initial beating up by the SNB in August, so when it finally announced the peg, it didn't have much work to do. Had it not beat up long CHF specs in August, the target would have required an incredible amount of purchases to implement. A central bank that lacks credibility can't just announce and expect things to magically fall into place.

Fast forward to the present and something odd is happening. As the chart below shows, the SNB is buying up huge amounts of euros in order to defend the CHF peg.

According to the Chuck Norris theory of central banking, this shouldn't be happening. A central bank that announces an explicit target, as the SNB has done, shouldn't have to expend any effort to protect that peg. But SNB foreign currency holdings have increased from under CHF250b to over CHF350 in just two months.

The SNB shouldn't have to defend the peg because in promising to purchase infinite amounts of euro deposits at 1.2000 EUR/CHF, private European banks will in turn step in and purchase euro deposits with CHF 1.2000. That's because with the SNB backstopping 1.2000,  they know that they can't lose by purchasing all the euros offered at that rate. Indeed, other banks will be willing to purchase euros for CHF 1.2001, knowing that the first group of private banks will in turn step in at 1.2000 because the SNB in turn has committed to stepping in at 1.2000. In sum, by simply expressing a commitment to buy euro deposits at 1.2000, the SNB creates a self-fulfilling loop whereby others defend the peg on the SNB's behalf by buying all euros at some amount greater than 1.2000. That's Chuck Norris in action.

The fact that the SNB has been forced to purchase large amounts of euro deposits indicates that for some reason, private banks haven't been stepping in to do their part. As a result, the SNB has had to fill the gap they have vacated. This could be because the banks don't believe that the peg is credible. Or maybe the SNB was never Chuck Norris to begin with. If the SNB isn't, then why would the ECB or Fed be Chuck Norris-like?

Saturday, August 11, 2012

Decoding Glasner on reflux, inside and outside money, and reflux

I had a few comments on a recent David Glasner post. Basically, I was trying to understand the way he reconciles various aspects of the monetary system, namely, inside and outside money, the arbitrage mechanism that links these two assets, and the price level.

David responded to me-
Inside money cannot trade at a discount relative to outside money because inside money is issued on the condition of its being convertible into outside money, so they always are exchangeable at par. If too much inside money is created (i.e., more than the public desires to hold given the relative attractiveness of holding inside money relative to alternatives including outside money) it refluxes back to the issuing banks. 
David says that excess inside money (say convertible bank notes) will reflux back to an issuing bank. But the only way this can happen, as far as I can see, is if somehow that bank's inside money trades at a slight discount to outside money (gold). David in his first sentence above says that inside money cannot trade at a discount. But how else can a reflux process emerge if one can't fall to a discount with the other?

So a temporary price discrepancy is necessary to enforce reflux and keep the quantity of outside money equal to demand. But what happens if inside money - say convertible bank notes - and outside money - fiat notes - are considered perfect substitutes by their users? After all, they can both be used to pay taxes, buy stuff, and "store value" over time.

David, for instance, points out that-
Because inside money and outside money are fairly close substitutes, the value of outside money is determined simultaneously in the markets for inside and outside money, just as the value of butter is determined simultaneously in the markets for butter and margarine. 
The problem here is... if inside and outside money are perfect substitutes, then given an excess issuance of convertible bank notes by a bank, why would the price discrepancy between inside and outside money that is necessary to drive reflux ever arise to begin with?

Rather, in an effort on the part of individuals and firms to rid themselves of their extra balances (either inside or outside money, they are indifferent) they will spend away both willy nilly. In spending away outside money, they will cause the price level to increase (the value of outside money to fall). David points this out:
If, however, the quantity of inside money increases because the public wants to hold the additional balances and are induced to hold inside money instead of outside money, then the value of outside money will tend to fall (causing the value of inside money to fall as well) unless the value of outside money is maintained by some form of convertibility or a price rule.
But doesn't this mean that issuing banks might cause infinite inflation by issuing inside money no one wants, thereby confirming Milton Friedman's wariness of free banking?

It would if currency users did in fact treat inside and outside money as equal.

But we live in a competitive banking system in which multiple banks issue inside money and receive other bank's inside money via cheque deposits and money transfers. Furthermore, currency users do not have uniform views about the quality of various money-like assets. Unlike individuals, competitive banks are picky and prefer to hold outside money rather than another bank's inside money. Put differently, banks are very sensitive to the store-of-value nature of inside money... they tend to view it as a financial asset characterized by risk and return, and not as a medium of exchange, and therefore view inside money as inferior to outside money due to its riskiness. Furthermore, holding another bank's inside money because they value its liquidity would be silly, since the bank already has its own liquidity factory. Thus the moment they receive another bank's inside money, a bank returns it to that issuer as fast as they can, settling with outside money. A reflux mechanism is thereby enforced by interbank settlement.

In sum, excess issuance of inside money by banks will not cause the price level to rise - it will cause the inside money to return to issuer.

Monday, August 6, 2012

The Interdistrict Settlement Account finally settles

Last year and earlier this year, debate concerning the Target2 mechanism in Europe and its growing imbalances shed some light on the equivalent institution in the US, the Interdistrict Settlement Account (ISA). The argument has been made to import the ISA structure into Europe, in particular, the yearly settling of accounts between district Reserve banks. If the ECB were to operate this way, the idea goes, then there would be some sort of quid pro quo on Target2 imbalances - European national central banks (NCBs) would not be able to accumulate debts to each other ad infinitum.

The observation was made that contrary to what one might expect, the ISA at the time did not seem to be balancing. Thus the idea that adoption of the Fed model would impose "firm limits" on intra-Eurosystem credit is invalid, since the Fed doesn't seem to always impose settlement on district Reserve banks. This April, though, the ISA seems to have been settled, as the charts below will illustrate. Nevertheless, the point still stands that the Federal Reserve System may choose to constrain itself by requiring interdistrict settlement or it may choose to forego that restraint.