Medieval thinkers were tempted to believe that if you throw a rock it flies straight until it runs out of force, and then it falls straight down. Economists are tempted to think of prices as a linear function of the “money supply”, and interest rates to be based on “inflation expectations”, which is to say expectations of rising prices.
The medieval thinkers, and the economists are “not even wrong”, to borrow a phrase often attributed to physicist Wolfgang Pauli. Science has to begin by going out to reality and observing what happens. Anyone can see that in reality, these tempting assumptions do not fit what occurs.
In my series of essays on interest rates and prices, I argued that the system has positive feedback and resonance, and cannot be understood in terms of a linear model. When I began this series of papers, the rate of interest was still falling to hit a new all-time low. Then on May 5,2013, it began to shoot up. It rose 83% over a period of exactly four months. That may or may not have been the peak (it has subsided a little since then).
Several readers asked me if I thought this was the beginning of a new rising cycle, or if I thought this was the End (of the dollar). As I expressed in Part VI, the End will be driven by the withdrawal of the gold bid on the dollar. Since early August, gold has become more and more abundant in the market. I think it is safe to say that this is not the end of the dollar, just yet. The hyperinflationists’ stopped clock will have to remain wrong a while longer. I said that the rising rate was a correction.
I am quite confident of this prediction, for all the reasons I presented in the discussion of the falling cycle in Part V. But let’s look at the question from a different perspective, to see if we end up with the same conclusion.
In the gold standard, the rate of interest is the spread between the gold coin and the gold bond. If the rate is higher, that is equivalent to saying that the spread is wider. If the rate is lower, then this spread is narrower.
A wider spread offers more incentive for people to straddle it, an act that I define as arbitrage. Another way of saying this is that a higher rate offers more incentive for people to dishoard gold and lend it. If the rate falls, which is the same as saying if the spread narrows, then there is less incentive and people will revert to hoarding to avoid the risks and capital lock-up of lending. Savers who take the bid on the interest rate (which is equivalent to taking the ask on the bond) press the rate lower, which compresses the spread.
It goes almost without saying, that the spread could never be compressed to zero (by the way, this is true for all arbitrage in all free markets). There are forces tending to compress the spread, such as the desire to earn interest by savers. But the lower the rate of interest, the stronger the forces tending to widen the spread become. These include entrepreneurial demand for credit, and most importantly the time preference of the saver—his reluctance to delay gratification. There is no lending at zero interest and nearly zero lending at near-zero interest.
I emphasize that interest is a spread to put the focus on a universal principle of free markets. As I stated in my dissertation:
“All actions of all men in the markets are various forms of arbitrage.”
Arbitrage compresses the spread that is being straddled. It lifts up the price of the long leg, and pushes down the price of the short leg. If one buys eggs in the farm town, then the price of eggs there will rise. If one sells eggs in the city center, then the price there will fall.
In the gold standard, hoarding tends to lift the value of the gold coin and depress the value of the bond. Lending tends to depress the value of the coin and lift the value of the bond. The value of gold itself is the closest thing to constant in the market, so in effect these two arbitrages move the value of the bond. How is the value of the bond measured—against what is it compared? Gold is the unit of account, the numeraire.
The value of the bond can move much farther than the value of gold. But in this context it is important to be aware that gold is not fixed, like some kind of intrinsic value. An analogy would be that if you jump up, you push the Earth in the opposite direction. Its mass is so heavy that in most contexts you can safely ignore the fact that the Earth experiences an equal but opposite force. But this is not the same thing as saying the Earth is fixed in position in its orbit.
The regime of irredeemable money behaves quite differently than the gold standard (notwithstanding frivolous assertions by some economists that the euro “works like” the gold standard). The interest rate is still a spread. But what is it a spread between? Does arbitrage act on this spread? Is there an essential difference between this and the arbitrage in gold?
Analogous to gold, the rate of interest in paper currency is the spread between the dollar and the bond. There are a number of differences from gold. Most notably, there is little reason to hold the dollar in preference to the government bond. Think about that.
In the gold standard, if you don’t like the risk or interest of a bond, you can happily hold gold coins. But in irredeemable paper currency, the dollar is itself a credit instrument backed by said government bond. The dollar is the liability side of the Fed’s balance sheet, with the bond being the asset. Why would anyone hold a zero-yield paper credit instrument in preference to a non-zero-yield paper credit instrument (except as speculation—see below)? And that leads to the key identification.
The Fed is the arbitrager of this spread!
The Fed is buying bonds, which lifts up the value of the bond and pushes down the interest rate. Against these new assets, the Fed is issuing more dollars. This tends to depress the value of the dollar. The dollar has a lot of inertia, like gold. It has extremely high stocks to flows, like gold. But unlike gold, the dollar’s value does fall with its quantity (if not in the way that the quantity theory of money predicts). Whatever one might say about the marginal utility of gold, the dollar’s marginal utility certainly falls.
The Fed is involved in another arbitrage with the bond and the dollar. The Fed lends dollars to banks, so that they can buy the government bond (and other bonds). This lifts the value of the bond, just like the Fed’s own bond purchases.
Astute readers will note that when the Fed lends to banks to buy bonds, this is equivalent to stating that banks borrow from the Fed to buy bonds. The banks are borrowing short to lend long, also called duration mismatch.
This is not precisely an arbitrage between the dollar and the bond. It is an arbitrage between the short-term lending and long-term bond market. It is the spread between short- and long-term interest rates that is compressed in this trade.
One difference between gold and paper is that, in paper, there is a central planner who sets the short-term rate by diktat. Since 2008, Fed policy has pegged it to practically zero.
This makes for a lopsided “arbitrage”, which is not really an arbitrage. One side is not free to move, even the slight amount of a massive object. It is fixed by law, which is to say, force. The economy ought to allow free movement of all prices, and now one point is bolted down. All sorts of distortions will occur around it as tension builds.
I put “arbitrage” in scare quotes because it is not really arbitrage. The Fed uses force to hand money to those cronies who have access to this privilege. It is not arbitrage in the same way that a fence who sells stolen goods is not a trader.
In any case, the rate on the short end of the yield curve is fixed near zero today, while there is a pull on the long bond closer to it. Is there any wonder that the rate on the long bond has a propensity to fall?
Under the gold standard, borrowing short to lend long is certainly not necessary  However, in our paper system, it is an integral part of the system, by its very design.
The government offers antiseptic terms for egregious acts. For example, they use the pseudo-academic term “quantitative easing” to refer to the dishonest practice of monetizing the debt. Similarly, they use the dry euphemism “maturity transformation” to refer to borrowing short to lend long, i.e. duration mismatch. Perhaps the term “transmogrification” would be more appropriate, as this is nothing short of magic.
The saver is the owner of the money being lent out. It is his preference that the bank must respect, and it is for his benefit that the bank lends. When the saver says he may want his money back on demand, and the bank presumes to lend it for 30 years, the bank is not “transforming” anything except its fiduciary duty, its integrity, and its own soundness. Depositors would not entrust their savings to such reckless banks, without the soporific of deposit insurance to protect them from the consequences.
Under the gold standard, this irrational practice would exist on the fringe on the line between what is legal and what is not (except for the yield curve specialist, a topic I will treat in another paper), a get-rich-quick scheme—if it existed at all (our jobs as monetary economists are to bellow from the rooftops that this practice is destructive).
Today, duration mismatch is part of the official means of executing the Fed’s monetary policy.
I have already covered how duration mismatch misallocates the savers’ capital and when savers eventually pull it back, the result is that the bank fails. I want to focus here on another facet. Pseudo-arbitrage between short and long bonds destabilizes the yield curve.
By its very nature, borrowing short to lend long is a brittle business model. One is committed to a long-term investment, but this is at the mercy of the short-term funding market. If short-term rates rise, or if borrowing is temporarily not possible, then the practitioner of this financial voodoo may be forced to sell the long bond.
The original act of borrowing short to lend long causes the interest rate on the long bond to fall. If the Fed wants to tighten (not their policy post-2008!) and forces the short-term rate higher, then players of the duration mismatch game may get caught off guard. They may be reluctant to sell their long bonds at a loss, and hold on for a while. Or for any number of other proximate causes, the yield curve can become inverted.
Side note: an inverted yield curve is widely considered a harbinger of recession. The simple explanation is that the marginal source of credit in the economy is suddenly more expensive. This causes investment in everything to slow.
At times there is selling of the short bond, at times aggressive buying. Sometimes there is a steady buying ramp of the long bond. Sometimes there is a slow selling slide that turns into an avalanche. The yield curve moves and changes shape. As with the rate of interest, the economy does best when the curve is stable. Sudden balance sheet stress, selloffs, and volatility may benefit the speculators of the world, but of course, it can only hurt productive businesses that are financing factories, farms, mines, and hotels with credit.
Earlier, I referred to the only reason why someone would choose to own the Fed’s liability—the dollar—in preference to its asset. Unlike with gold, hoarding paper dollar bills serves no real purpose and incurs needless risk of loss by theft. The holder of dollars is no safer. He avoids no credit risk; he is exposed to the same risk as is the bondholder is exposed. The sole reason to prefer the dollar is speculation.
As I described in Theory of Interest and Prices in Paper Currency, the Fed destabilizes the rate of interest by its very existence, its very nature, and its purpose. Per the above discussion, the Fed and the speculators induce volatility in the yield curve, which can easily feed back into volatility in the underlying rate of interest.
The reason to sell the bond is to avoid losses if interest rates will rise. Speculators seek to front-run the Fed, duration mismatchers, and other speculators. If the Fed will “taper” its purchase of bonds, then that might lead to higher interest rates. Or at least, it might make other speculators sell. Every speculator wants to sell first.
Consider the case of large banks borrowing short to lend long. Let’s say that you have some information that their short-term funding is either going to become much harder to obtain, or at least significantly more expensive. What do you do?
You sell the bond. You, and many other speculators. Everyone sells the bond.
Or, what if you have information that you think will cause other speculators to sell bonds? It may not even be a legitimate factor, either because the rumor is untrue (e.g. “the world is selling Treasury bonds”) or because there is no valid economic reason to sell bonds based on it.
You sell the bond before they do, or you all try to sell first.
I have been documenting numerous cases in the gold market where traders use leverage to buy gold futures based on an announcement or non-announcement by the Fed. These moves reverse themselves quickly. But no one, especially if they are using leverage, wants to be on the wrong side of a $50 move in gold. You sell ahead of the crowd, and you buy ahead of the crowd. And they try to do it to you.
I think it is likely that one of these phenomena, or something similar, has driven the rate on the 10-year Treasury up by 80%.
I would like to leave you with one take-away from this paper and one from my series on the theory of interest and prices. In this paper, I want everyone to think about the difference between the following two statements:
- The dollar is falling in value
- The rate of interest in dollars must rise
It is tempting to assume that they are equivalent, but the rate of interest is purely internal to the “closed loop” dollar system. Unlike a free market, it does not operate under the forces of arbitrage. It operates by government diktats, and hordes of speculators feed on the spoils that fall like rotten food to the floor.
From my entire series, I would like the reader to check and challenge the sacred-cow premises of macroeconomics, the aggregates, the assumptions, the equations, and above all else, the linear thinking. I encourage you to think about what incentives are offered under each scenario to the market participants. No one even knows the true value of the monetary aggregate and there is endless debate even among economists. The shopkeeper, miner, farmer, warehouseman, manufacturer, or banker is not impelled to act based on such abstractions.
They react to the incentives of profit and loss. Even the consumer reacts to prices being lower in one particular store, or apples being cheaper than pears. If you can think through how a particular market event or change in government policy will remove old incentives and offer new incentives, then you can understand the likely first-order effects in the market. Of course each of these effects changes still other incentives.
It is not easy, but this is the approach that makes economics a proper science.
P.S. As I do my final edits on this paper (October 4, 2013), there is a selloff in short US T-Bills, leading to an inversion at the short end of the yield curve. This is due, of course, to the possible effect of the partial government shutdown. The government is not going to default. If this danger were real, then there would be much greater turmoil in every market (and much more buying of gold as the only way to avoid catastrophic losses). The selloff has two drivers. First, some holders of T-Bills need the cash on the maturity date. They would prefer to liquidate now and hold “cash” rather than incur the risk that they will not be paid on the maturity date. Second, of course speculators want to front-run this trade. I put “cash” in scare quotes because dollars in a bank account are the bank’s liability. The bank will not be able to honor this liability if its asset—the US Treasury bond—defaults. The “cash” will be worthless in the very scenario that bond sellers are hoping to avoid by their very sales. When the scare and the shutdown end, then the 30-day T-Bill will snap back to its typical rate near zero. Some clever speculators will make a killing on this move.
 See the Monetary Metals Supply and Demand Report
 I argue that it always fails in the end in Duration Mismatch Always Fails