Sunday, November 23, 2014

Not your father's price index: the Billion Prices Project

The price of 52 Samsung TVs gathered by the BPP, April 2008 - November 2009 (Cavallo)

In a previous post, I mentioned that the Billion Prices Project (BPP) contradicts the claims of those who believe that the government understates inflation data. The BPP crawls major US retailers' websites and scrapes them for price data, compiling an overall US Daily Index that is available on its website. The deviation between this index and the official CPI is minimal, as the above link shows.

The BPP isn't your father's price index—it shouldn't be viewed as a perfect substitute for the CPI. So use it wisely. What follows are a few details that I've gleaned from several papers on the topic of online price indexes as well my correspondence with Roberto Rigobon, one of the project's founders.

The most obvious difference between it and the CPI is in the datasets:

1) Online vs offline: The price data to generate the CPI is harvested by Bureau of Labour Statistics (BLS) inspectors who trudge through brick & mortar retailers. Rigobon and his co-founder Alberto Cavallo get their data by sending out lightning fast algorithms to scrape the websites of online retailers.

2) Wide vs Narrow: BLS inspectors compile prices on a wide range of consumer goods and services. According to Cavallo, only 60% of the items that are in the CPI are available online. The ability to track service prices online is particularly limited given the fact that most large retailers websites only sell goods.

Let's get into some more specifics about what is included in the BPP, because there seems to be some confusion about this in the online discussion. Some commentators have mentioned that the BPP doesn't include gasoline prices. Rigobon informs me that this is wrong, gas prices are included in the US Daily Index. As for the cost of housing, my understanding is the BPP does track real estate data. It incorporates these prices using the same methodology as the BLS. So any deviation between the BPP and CPI should not be attributed to the BPP's lack of either gas prices or housing.

Lastly, despite the fact that service prices are under-represented online, the BPP's US Daily Index does include a number of services. According to Rigobon, the easiest ones to track are things like health insurance, transportation, restaurants, hotel, and haircuts. Others are hard to track, like the cost of public education. My understanding is that Rigobon and Cavallo may use proprietary methods to calculate service prices by referring to various goods' prices as proxies (see here). For instance, in this BIS comment on the BPP, it is noted that the price of education can be computed from prices of text books, uniforms, energy and construction materials, all of which represent 75% of cost of education.

3) Often vs rare: The BPP's algorithms trawl retailer websites every day. BLS inspectors stroll through the malls just once each month.

Another big difference is in the publication of the data:

4) Now vs later: The BPP is reported three days after the data has been gathered, and ten days for non-subscribers. CPI is reported with a long delay, usually the second or third week following the month being covered.

The next few differences are a little more technical:

5) Fixed vs Responsive: Both indexes measure entirely different consumption baskets. The BLS surveys U.S. households every few years in order to gather information about their spending habits. It uses this information to construct a fixed representative basket of goods & services consumed by Americans, and then proceeds to fill in the data each month. This survey approach results in a CPI basket that takes time to adjust to new products. Should a revolutionary device, say a universal mind reader, suddenly becomes popular, it won't be reflected in the CPI till the next survey.

Think of the BPP as capturing a dynamic market-determined consumption basket. The BPP basket is comprised of whatever goods retailers happen to be selling online that day in order to meet customer demand. Because retailers are constantly updating their websites, August 7's basket could be different from August 8's. This means that new goods will be quickly incorporated into Rigobon and Alvarez's inflation calculation. In other words, when universal mind readers do catch on, the BPP will incorporate this data way before the BLS will.

One of the most interesting differences is the difference in methodology:

6) Small vs Large Sample size: The BLS delicately samples offline prices whereas the BPP bulldozes through a large percentage of the entire population of online retailers' prices.

There are millions of goods sold in the US, and it would be cruel to force BLS inspectors to collect prices for all of them. To simplify the calculation, the BLS brain trust chooses individual products to serve as ideal representatives for given product categories. Take dishwashers. To represent the category, they might select the Whirlpool WTD-10 or some such model. A BLS data collector in New York City will go every month to a specific store, say Macy's on West 34th St, and grab that specific model's price. The repetitive use of the same product and location ensures that the New York City dishwasher price index is not corrupted by changes that have little to do with purchasing power. (The alternating collecting of prices from Macy's on West 34th and Nordstrom's on Union Square might introduce price changes having little to do with inflation.)

Because their algorithms are whip fast and don't require salaries, Alvarez and Rigobon can afford to send them out each day to Macy's website to gather the price of every single dishwasher. They do this for each of the major online retailers, say Walmart, Target, and Best Buy. The final assemblage of prices represents something close to the entire population of online US dishwasher prices on every single day!

This segues into the thorny problem of adjusting for quality changes. They both use different techniques:

7) Statistical vs market-based quality adjustments: As I pointed out, the BLS samples one good to represent a given category rather than canvassing the full range of products within that category. This causes some difficulties when that one good is replaced by another product.

Let's return to the Macy's example. Say Macy's stops stocking the Whirlpool WTD-10. On arriving at Macy's a few weeks later, our flummoxed BLS data collector has to find a replacement in order to keep the dishwasher price category up to date. Let's say she grabs the price of a General Electric XK-400 from across the aisle. The GE is priced $50 higher than the missing Whirlpool was during the inspector's previous visit. The problem is this: how does the BLS determine how much of that $50 increase is due to changes in quality and how much is due to changes in inflation? If the GE is the same in every way to the Whirlpool except its boasts a turbo wash option, then some portion of the $50 increase is due to the higher quality of the GE. But how much?

Because Cavallo and Rigobon's tireless algorithms regularly retrieve multiple product prices for each category rather than single monthly representatives, they can use the overlapping nature of the data to seamlessly splice in new products. Let's say that the expensive GE dishwasher is introduced to Macy's website. It is sold on the same page as the existing and cheaper Whirlpool for a few days at which point the latter is removed. On the day the GE first appears, the BPP ascribes its higher price to its superior quality. While the GE drives up the average price of dishwashers on Macy's dishwasher page, the purchasing power of a Macy's shopper hasn't been altered, rather, a given dollar buys more 'dishwashing services' than before. Only on day 2, after the GE's price has been retrieved a second time by the algorithms, is it allowed to start affecting the index, since any price change thereafter is considered to be due to inflation, not quality.

The assumption that the GE's premium is due entirely to quality is based on the idea that market prices are accurate measures of all that is known by producers and consumers about a given set of products.

Because CPI collectors have limited resources and typically only collect the price of one representative dishwasher, they usually can't rely on the overlap between dishwasher model prices to measure quality changes. One method they have developed to compute quality changes is hedonic regression. In brief, a dishwasher is conceptually broken up into a package of characteristics, including its size, time per run, energy efficiency, etc. When the Whirlpool is suddenly dropped by Macy's and the GE added, CPI data collectors try to determine what sorts of new characteristics have been incorporated in the GE and then use regression methods to determine the dollar value of that characteristic.

So to sum up, to calculate quality changes, the BPP piggy backs on the power of the market to price difference in quality. The BLS uses econometric methods (among other tools) to control for quality changes.

Here is a big one, the difference in ownership of the indexes:

8) Private vs public: The CPI is compiled by the BLS and funded by taxpayers, whereas Rigobon and Cavallo have incorporated a private company called Pricestats to compile the BPP US Daily index and its many other indexes. PriceStats work in partnership financial-giant State Street to distribute the data to paying subscribers.

Which leads into the last major difference:

9) Transparent vs opaque: There is loads of documentation on the CPI. If you have any questions, call up the BLS and a researcher will walk you through it—it's your right as a taxpayer. PriceStats can only reveal so much information because their methods are proprietary (although Dr Rigobon was kind enough to answer a number of my questions). I suspect they are hesitant to reveal too much of information because the retailers on which they have gathered data might view this as a potentially threatening action. Not so with the CPI.

So those are some of the features of each index. In the case of the BPP, the difficulty of getting public information on their methodology is probably the biggest bug, although the founders are forthcoming on general questions. Maybe if national statistics agencies start adopting BPP data collection methods, the transparency problem will be solved, since public agencies have no competitive reasons (and less legal ones) to hold back information on methodology. There seem to be rumblings in this direction: Statistics New Zealand says that they are in the early stages of a collaboration with with PriceStats to develop online price indexes (link).

For now the public is lucky to get access to the US Daily Index, even on a 10-day delay. When CPI numbers are reported the bond market quakes. For hedge funds, getting a hint of what the upcoming government inflation print will be before anyone else is probably worth a lot of money. No doubt that's why they are willing to pay to subscribe to get PriceStat's numbers. These funds would probably prefer if the public were not privy to the US Daily Index as it reduces the information's value. The amount they'd be willing to pay PriceStats to yank the US Daily Index from the public domain would be a good indicator of the value the public gains by getting free access to it. It could be a substantial number.


Back to initial reason for writing about the BPP; the gold bugs (Gulp, you thought I'd forgotten about you, right?). Your typical gold bug will sagely mention some esoteric price that has risen at an incredible rate over the last few years, like the price of shitaake mushrooms or a 1982 GI-Joe Snake Eyes collectors action figure (here is Peter Schiff using the Big Mac). A gold bug is convinced that their preferred data series is sufficiently strong evidence to justify declaring inflation to be stratospheric and the entire CPI null and void.

What makes gold bugs think that their one or two pet prices are a superior measure of the dollar's purchasing power than the BPP US Daily Index? Crickets. That sums up the gold bug response to the BPP's existence. If not crickets, then desperate attempts to change the subject.

Gold bugs don't like to talk about the BPP because they don't want to be dissuaded from their views—they find too much comfort in them. With the BPP continuing to move in line with the CPI, the gold bug community's cognitive dissonance is growing. At some point, the squirm-level will get large enough that they'll have to do something about it. No doubt the easiest route will be to come up with a fiction that discredits the BPP US Daily index. Well, hey gold bugs, here's a conspiracy theory you can use to save yourselves some painful cognitive dissonance... the Billion Prices Index went offline for a period of time, just when it appeared to be showing a break with the CPI index. When it went back online, the two started to converge. Could it be that Rigobon and Alvarez were brought into some FBI dungeon and re-programmed, the BPP moving more in line with the party line after they emerged? Yeah, that's it.

Saturday, November 15, 2014

Sign Wars

Does a lowering of a central bank's interest rates create inflation or deflation? Dubbed the 'Sign Wars' by Nick Rowe, this has been a recurring debate in the economics blogosphere since at least as far back as 2010.

The conventional view of interest rate policy is that if a central bank keeps its interest rate too low, the inflation rate will steadily spiral higher. Imagine a cylinder resting on a flat plane. Tilt the plane in one direction —a motif to explain a change in interest rates—and the cylinder, or the price level, will perpetually roll in the opposite direction, at least until the plane's tilt (i.e. the interest rate) has been shifted enough in a compensatory way to halt the cylinder's roll. Without a counter-balancing shift, we get hyperinflation in one direction, or hyperdeflation in the other.

The heretical view, dubbed the Neo-Fisherian view by Noah Smith (and having nothing to do with Irving Fisher), is that in response to a tilt in the plane, the cylinder rolls... but uphill. Specifically, if the interest rate is set too low, the inflation rate will jump either instantaneously or more slowly. But after that, a steady deflation will set in, even without the help of a counter-balancing shift in the interest rate. We get neither hyperinflation nor hyperdeflation. (John Cochrane provides a great introduction to this viewpoint).

Many pixels have already been displayed on this subject, about the only value I can add is to translate a jargon-heavy academic debate into a more finance-friendly way of thinking. Let's approach the problem as an exercise in security analysis.

First, we'll have to take a detour through the bond market, then we'll return to money. Consider what happens if IBM announces that its 10-year bond will forever cease to pay interest, or a coupon. The price of the bond will quickly plunge. But not forever, nor to zero. At some much lower price, value investors will bid for the bond because they expect its price to appreciate at a rate that is competitive with other assets in the economy. These expectations will be motivated by the fact that despite the lack of coupon payments, the bond still has some residual value; specifically, IBM promises a return of principal on the bond's tenth year.

Now there's nothing controversial in what I just said, but note that we've arrived at the 'heretical' result here. A sudden setting of the interest rate at zero results in a rapid dose of inflation (a fall in the bond's purchasing power) as investors bid down the bond's price, followed by deflation (a steady expected rise in its value over the next ten years until payout) as its residual value kicks in. The bond's price does not "roll" forever down the tilted plane.

Now let's imagine an IBM-issued perpetual bond. A perpetual bond has no maturity date which means that the investor never gets their principle back. Perpetuals are not make-believe financial instruments. The most famous example of perpetual debt is the British consol. A number of these bonds float around to this day after having been issued to help pay for WWI. When our IBM perpetual bond ceases to pay interest its price will quickly plunge, just like a normal bond. But it's price won't fall to zero. At some very low level, value investors will line up to buy the bond because its price is expected to rise at a competitive rate. What drives this expectation? Though the bond promises neither a return of principal nor interest payments, it still offers a fixed residual claim on a firm's assets come bankruptcy, windup, or a takeover. This gives value investors a focal point on which they can price the instrument.

So with a non-interest paying perpetual bond, we still get the heretical result. In response to a plunge in rates, we eventually get long term deflation, or a rise in the perpetual's price, but only after an initial steep fall.  As before, the bond's price does not fall forever.

Now let's bring this back to money. Think of a central bank liability as a highly-liquid perpetual bond (a point I've made before). If a central banker decides to set the interest rate on central bank liabilities at zero forever, then the purchasing power of those liabilities will rapidly decline, much like how the cylinder rolls down the plane in the standard view. However, once investors see a profit opportunity in holding those liabilities due to some remaining residual value, that downward movement will be halted... and then it will start to roll uphill. Once again we get the heretical result.

The residual claim that tempts fundamental investors to step in and anchor the price of a 0% yielding central bank liability could be some perceived fixed claim on a central bank's assets upon the bank's future dissolution, the same feature that anchored our IBM perpetual. Or it could be a promise on the part of the government to buy those liabilities back in the future with some real quantity of resources.

However, if central bank liabilities don't offer any residual value whatsoever, then we get the conventional result. The moment that the central bank ceases to pay interest, the purchasing power of a central bank liability declines...forever. Absent some residual claim, no value investor will ever step in and set a floor. In the same way, should an IBM perpetual bond cease to pay interest and it also had all its residual claims on IBM's assets stripped away, value investors would never touch it, no matter how low it fell.

So does central bank money boast a residual claim on the issuer? Or does it lack this residual claim? The option you choose results in a heretical result or a conventional result.


What does the data tell us, specifically the many cases of hyperinflation? As David Beckworth has pointed out, the conventional explanation has no difficulties explaining the Weimar hyperinflation; the Reichsbank kept the interest rate on marks fixed at very low levels between 1921 and 1923 so that the price level spiraled ever upwards. Heretics seem to have difficulties with Weimar—the deflation they predict never set in.

Here's one way to get a heretical explanation of the Wiemar inflation. Let's return to our analogy with bonds. What would it take for the price of an IBM perpetual bond to collapse over a period of several years, even as its coupon rate remained constant? For that to happen, the quality of the bond's perceived residual value would have to be consistently deteriorating. Say IBM management invested in a series of increasingly dumb ventures, or it faced a string of unbeatable new competitors entering its markets. Each hit to potential residual value would cause fundamental investors to mark down IBM's bond price, even though the bond's coupon remained fixed.

Now assuming that German marks were like IBM perpetual bonds, it could be that from 1921 to 1923, investors consistently downgraded the value of the residual fixed claim that marks had upon the Reichsbank's assets. Alternatively, perhaps the market consistently reduced its appraisal of the government's ability to buy marks back with real resources. Either assumption would have created a consistent decline in the purchasing power of marks while the interest rate paid on marks stayed constant.

Compounding each hit to residual value would have been a decline in the mark's liquidity premium. When the price of a highly-liquid item begins to fluctuate, people ditch that item for competing liquid items with more stable values. With less people dealing in that item, it becomes less liquid, which reduces the liquidity premium it previously enjoyed. This causes the item's purchasing power to fall even more, forcing people to once again turn to alternatives, thus making it less liquid and igniting another round of cuts to its liquidity premium and therefore its price, etcetera etcetera. In Weimar's case, marks would have been increasingly replaced by dollars and notgeld.

So consistent declines in the mark's perceived residual value, twinned with a shrinking in its liquidity premium, might have been capable of creating a Weimar-like inflation, all while the Reichsbank kept its interest rate constant.


That's not to say that central bank liabilities do have a residual value and that the heretical result is necessarily the right one. Both possibilities make sense, and both can explain hyperinflations. But to determine which is right, we need to go in and do some gritty security analysis to isolate whether central bank money possesses a fixed residual claim on either central bank assets or future government resources. Parsing the fine print in central bank acts and government documents to tease out this data is the task of lawyers, bankers, historians, fixed income analysts, and accountants. And they would have to do a separate analysis for each of the world's 150 or so central banks and currencies, since each central bank has its own unique constituting documents. In the end we might find that some currencies are conventional and others are heretic, so that some central banks should be running conventional monetary policies, and others heretic policies. 

In closing, a few links. I've taken a shot at a security analysis of central bank liabilities in a number of posts (here | here | here), but I don't think that's the final word. And if you're curious how the Weimar inflation ended, go here.

Tuesday, November 4, 2014

Gilded cage

This blog wouldn't be around if it wasn't for gold bugs.

Many moons ago my former-employer (and friend), the truest gold bug you'd ever meet, would lecture everyone in the office for hours about imminent hyperinflation, the wonders of the gold standard, and why gold should be worth $10,000. Fascinated, but unsure what to make of his diatribes, I started to read about the history of monetary systems, all of which would eventually provide grist for this blog.

A gold bug will typically have the following characteristics. 1) An abnormally-sized portion of their investing portfolio will be allocated to the yellow metal; 2) they believe in an eventual 'day of reckoning' when gold's price rises into the stratosphere, the mirror image of which is hyperinflation; 3) their investing case for gold is twinned with strong moral view on the decrepitude of the current monetary system and/or society in general; and 4) they are 100% sure that the monetary system's collapse will lead to the flowering of a new and virtuous system, a gold standard.

One thing I discovered fairly early on from my interactions with the gold bug community is that there's no point in debating a gold bug. In any debate, you should be able to ask your opponent what evidence they'd accept as proving their idea to be wrong. Gold bugs are loathe to submit such a list. After all, to do so would open up the possibility that they might have to precommitt themselves to changing their mind, which is the last thing they want to do. A gold bug's ideas are comforting to them. They've structured their entire mental landscape around these ideas, not to mention their entire life's savings and often careers around them.

Gold bugs have a powerful set of defense mechanisms to protect their ideas from outside threat. These mechanism, I'll call them 'mental bodyguards', will kill on sight any idea or bit of evidence that runs contrary to the gold bug schema, thus saving the gold bug from the discomfort, and potential danger, of having to weigh each new bit of data on its own merit.

For instance, consider the fact that central bank money was unmoored from the gold peg in 1968 (almost 50 years ago!). The monkeys behind the wheel should have caused hyperinflation by now and all those financial Noahs who were smart enough to jump into the gold boat before the fiat flood should be fabulously wealthy. But gold trades at just $1200 or so, not far above $850 levels set in 1980. Except for a few exceptions like Zimbabwe, hyperinflation hasn't happened.

Gold bugs can rationalize this contradiction because they possess a 'mental bodyguard' that absolves them of any responsibility for the timing of their predictions. Like the Millerite movement—which predicted the second coming of Jesus Christ on March 21, 1884, only to have to push the date to April 18 when nothing happened, and when that day passed uneventfully, bumped the event to October 22—gold bugs can keep pushing the day-of-reckoning further into the future without suffering any mental dissonance. Using an even more impressive bit of mental-Aikido they turn disconfirmation into a positive. The longer gold's meteoric rise is forestalled, say gold bugs, the more time it provides true believers with an opportunity to accumulate a larger stash of the stuff.

Another powerful mental body guard is the invocation of "them". Gold bugs invariably blame vague external and impersonal forces for wreaking havoc on the noble intentions of gold bugs and the upwards trajectory of the metal's price. They  may be the Federal Reserve, the plunge protection team, or a cabal of Jewish bankers (politically-correct gold bugs just blame Goldman Sachs). When gold falls in price it's always because of the the machinations of these oppressors, without which the metal would be worth $12,000 or $13,000 by now. (Yes, gold bugs like to refer to gold as "the" metal, presumably to differentiate it from all the plebeian metals)

Thanks to the them mental body guard, the inability of gold bug predictions to be borne out in reality is never due to any inherent weakness in the ideas themselves, but to outside interference. Doubts are conveniently refocused on something external like Ben Bernanke and the Fed, upon which gold bugs regularly bestow two minute hates.

Other mental bodyguards that prove useful in protecting the core gold bug ideology include the knee jerk discredit that gold bugs level at both the economics profession and economic data. Gold bugs screen out economists by deriding them as mainstream and therefore (obviously!) puppets of the system. The shoot-first assumption of guilt spares gold bugs from having to engage with these economists' potentially contradictory ideas on a level playing field. The same goes for inflation data, which they dismiss out of hand as being 'cooked'. And if you try mentioning the MIT Billion Prices Index to them, they hum loudly and put their fingers in their ears. (Although when there's any sort of divergence between the BPI and CPI, they suddenly start to make noise).

The awful returns that gold and especially gold shares have provided over the decades have impoverished many gold bugs as well as those unlucky enough to listen to them. Yep, I've seen the year-end statements. Yet somehow the gold bug meme continues to limp on. That's because gold bugs are less concerned about making money than upholding "the cause", as they like to refer to it. The cause is a vague combination of the promotion of a gold standard and a +$10,000 gold price, where simply holding gold through all downturns is an expression of support for that cause. Mere financial losses cannot keep them down.

Now I've been tough on the gold bugs in this post, but the fact is that gold bugs would probably say that both myself and any of their many accusers harbour mental body guards of our own. And the gold bugs probably wouldn't be entirely wrong. With so much time and energy having been invested in the various things we know and believe, a bit of cognitive dissonance is only natural. I'd argue that the gold bugs having walked much further out along that plank than their critics.

This post won't change the minds of any gold bugs—as I already pointed out, they've made up their minds long ago. But if you're a busy individual with some money to invest, and you're considering a gold bug advisor, remember that the fate of your investment may take second seat to the gold bug's devotion to the cause. Be wary.

Tuesday, October 28, 2014

Fear of illiquidity

There are thousands of fears, from arachnophobia to globophobia (fear of balloons) to zoophobia (fear of animals). What might the fear of market illiquidity look like?

Say that you are petrified that a day will come when markets will be too illiquid for you to convert your wealth into the things you need. There are two ways for you to buy complete peace of mind.

The first strategy involves selling everything you own now and buying checking deposits, the sine qua non liquid asset. Get rid of the house, the bonds, the stocks, the car, your couch, and your books. Use some of the proceeds to rent a house and a car, borrow books, and lease furniture. In renting back the stream of consumption benefits that you've just sold, your level of consumption stays constant. Negotiate the rental arrangements so that the lessor—the owner—cannot cancel them, and so that you can walk out of them at a moment's notice. Structuring things this way ensures that rental obligations in no way inhibit your ability to stay liquid. With your hoard of highly liquid deposits and array of rental agreements, you've secured a state of perfect liquidity. Relax. Breathe in. Enjoy your life.

The second way to perfectly hedge yourself from illiquidity risk would be to buy liquidity insurance on everything you own. For instance, an insurer would guarantee to purchase your house whenever you want to sell at the going market price. Same with your stocks, and bonds, your car and couches and your books. With every one of your possessions convertible into clean cold cash upon a moment's notice thanks to the insurer, you can once again relax, put your legs up, and lean back on the couch.

Since both strategies lead you to the same infinitely liquid final resting place, arbitrage dictates that the cost of pursuing these two strategies should be the same. Consider what would happen if the liquidity insurance route was cheaper. All those desiring a state of infinite liquidity would clamor to buy insurance, pushing the price of insurance higher until it was no longer the better option. If the checking deposit/rental route was cheaper, then everyone would sell all their deposits and rent stuff, pushing rental prices higher until it was no longer the more cost-efficient option.

Now I have no idea what liquidity insurance should actually cost. But consider this: liquidity option #1 is a *very* expensive strategy. To begin with, you'd be forgoing all the interest and dividends that you'd otherwise be earning on your bonds and stocks. Checking deposits, after all, offer no interest. Compounded over many years, that comes out to quite a bit of forfeited wealth. Second, you'd have to rent everything. And the sort of rent you'd have to negotiate would be costlier than normal rent. Last time I checked, most landlords require several months notice before a renter can be released from their rental obligation. But the rental agreements you have negotiated require the owner to accept a return of leased property whenever *you* want—not when they want. And that feature will be a costly one.

Since option #1 is so expensive, arbitrage requires that option #2 will be equally expensive. Let's break it out. Option #2, liquidity insurance, allows you to keep the existing flows of income from stocks and bonds as well as saving you from the obligation of paying high rent (you get to keep your house and all the other stuff). Not bad, right? Which means that in order for you to be indifferent between option #1 and #2, the cost of insurance must be really really high. If it wasn't, everyone would choose to go the insurance route.

So who cares ? After all, liquidity insurance doesn't exist, right? Wrong. Central banks are significant providers of liquidity insurance. They insure private banks against illiquidity by promising to purchase bank assets at going market prices whenever the bank requires it. This isn't full and complete liquidity insurance— there are a few assets that even a central bank won't touch—but it's close enough.

The upshot is that banks are well-protected from illiquidity. They get to keep all their interest-yielding assets and at the same time can rest easy knowing that the central bank insures that those assets will always be as good as cash. Consider what things would be like for private banks if the central bank were to get out of the liquidity insurance business. Now, the only way for bankers to replicate central bank-calibre liquidity protection would be for them to pursue option #1: sell their loan books and bond portfolios for 0%-yielding cash. But then they'd be foregoing huge amounts of income. They might not even be profitable.

With logic dictating that the cost of buying liquidity insurance needs to be pretty high, are modern central banks charging sufficiently stiff rates on liquidity insurance? I'm pretty sure they aren't. Regular insurers like lifecos require periodic premium payments, even if the event that said insurance covers hasn't occurred. But the last time I read a bank annual report, there was no line item for liquidity insurance premiums. It seems to me, and I could be wrong, that central banks are providing liquidity insurance without requiring any sort of quid pro quo. Feel free to correct me in the comments section.

Say that I'm right and that central banks are providing private banks with underpriced liquidity insurance. Central banks are ultimately owned by the taxpayer, which means that taxpayers are providing private banks with artificially cheap liquidity insurance. And that's not a fair burden to put on them. Nor is the underpricing of insurance a good strategy, since it results in all sorts of institutions getting insurance when they don't necessarily deserve it.

Does anyone know if central banks have any sort of rigorous model for determining the price they charge for liquidity insurance. Or are they just winging it? ... it sure seems like it to me.

Sunday, October 19, 2014


Recent posts by Adrian Hope Baille and Sina Motamedi have got me thinking again about the idea of the Federal Reserve (or any other central bank for that matter) adopting bitcoin technology. Here's an older post of mine on the idea, although this post will take a different tack.

The bitcoin ethos enshrines the idea of a world free from the totalitarian control of central banks. So in exploring the idea of Fed-run bitcoin-style ledger, I realize that I run the risk of being cast as Darth Vader (or even *yikes* the Emperor) by bitcoin true believers. So be it. While I do empathize with the bitcoin ideal—I support freedom in banking—I rank the importance of bitcoin-as-product above bitcoin-as-philosophy. And at the moment, bitcoin is not a great product. While bitcoin has many useful features, these are all overshadowed by the fact that its price is too damn volatile for it to be be taken seriously as an exchange medium. This volatility arises because bitcoin lacks a fundamental value, or anchor, a point that I've written about many times in the past. However, there is one way to fix the crypto volatility problem...

Enter Fedcoin

Setting up the apparatus would be very simple. The Fed would create a new blockchain called Fedcoin. Or it might create a Ripple style ledger by the same name. It doesn't matter which. There would be an important difference between Fedcoin and more traditional cryptoledgers. One user—the Fed—would get special authority to create and destroy ledger entries, or Fedcoin. (Sina Motamedi gives a more technical explanation for how this would work in the case of a blockchain-style ledger)

The Fed would use its special powers of creation and destruction to provide two-way physical convertibility between both of its existing liability types—paper money and electronic reserves—and Fedcoin at a rate of 1:1. The outcome of this rule would be that Fedcoin could only be created at the same time that an equivalent reserve or paper note was destroyed and, vice versa, Fedcoin could only be destroyed upon the creation of a new paper note or reserve entry.

So unlike bitcoin, the price of Fedcoin would be anchored. Should Fedcoin trade at a discount to dollar notes and reserves, people would convert Fedcoin into these alternatives until the arbitrage opportunity disappears, and vice versa if Fedcoin should trade at a premium.

As for the supply of Fedcoin, it would effectively be left free to vary endogenously, much like how the Fed currently let's the market determine the supply of Fed paper money. This flexibility stands in contrast to the fixed supply of bitcoin and other cryptocoins. The mechanism would work something like this. Should the public demand Fedcoin, they would have to bring paper dollars to the Fed to be converted into an equivalent number of new Fedcoin ledger entries, the notes officially removed from circulation and shredded. As for banks, if they wanted to accumulate an inventory of Fedcoin, they would exchange reserves for Fedcoin at a rate of 1:1, those reserves being deleted from Fed computers and the coins added to the Fedcoin ledger.

Symmetrically, unwanted Fedcoin would reflux to the central bank in return for either newly-created cash (in the case of the public) or reserves (in the case of banks), upon which the Fed would erase those coins from the ledger. The upshot is that the Fed would have no control over the quantity of Fedcoin—it would only passively create new coin according to the demands of the public.

Apart from that, Fedcoin would be similar in nature to most other cryptoledgers. All Fedcoin transactions would be announced to a distributed network of listening nodes for processing and verification. In other words, these nodes, and not the Fed, would be responsible for maintaining the integrity of the Fedcoin ledger.

Why implement Fedcoin?

The main reasons that the Fed would implement Fedcoin would be to provide the public with an innovative and cheap payments option, and to provide the taxpayer with tax savings.

The public would enjoy all the benefits of bitcoin including fast transaction speeds, cheap transaction costs, and the ability to transact almost anywhere and with almost anyone as long as all parties to a transaction had a smartphone and the right software. At the same time Fedcoin's stability would immediately differentiate it from bitcoin. No longer would users have to fear losing 50% of their purchasing power prior to making a transaction.

Fedcoin's distributed architecture would be both complementary and in many ways superior to Fedwire, a centralized system which currently provides for the transferal of Fed electronic reserves among banks. I won't bother getting into the specifics: see this old post.

By introducing Fedcoin, the Fed would also lower its costs. While I haven't done the calculations, I have little doubt that running a distributed cryptoledger is far cheaper than maintaining billions of paper notes in circulation. Paper currency involves all sorts of outlays including designing and printing notes, collecting, processing and storing them, as well as constantly defending the note issue against counterfeiters. A distributed ledger does all this at a fraction of the cost. As Fedcoin begins to displace cash, and I think that this would steadily happen over time due to its superiority over paper, the Fed's costs would fall and its profits rise to the benefit of the taxpayer.

Fedcoin would have no impact on monetary policy

Fed officials might balk at giving the idea a shot if they feared that adopting a Fed cryptoledger would impede the smooth functioning of Fed monetary policy. They needn't worry.

The Fed currently exercises control over the price level by varying the quantity of reserves and/or the interest paid on reserves. The existence of cash doesn't get in the way of this process, nor has it ever gotten in the way. Bringing in a third liability type, Fedcoin, the quantity of which is designed to fluctuate in the same way as cash, would likewise have no impact on monetary policy. The Fed would continue to lever the return on reserves in order to get a bite on prices while allowing the market to independently choose the quantity of Fedcoin and cash it wished to hold.

Well, almost none: Interest on Fedcoin and the zero lower bound

Ok, I sort of lied in the last paragraph. While it happens only rarely, there are times when cash does get in the way of monetary policy, and so would Fedcoin if it were implemented. If the Fed needs to reduce rates on reserves to negative levels in order to hit its price and employment targets, the existence of cash impedes the smooth slide below zero. With reserves yielding -2% and paper notes yielding 0%, reserves would quickly be converted en masse into cash until only the latter remains. At that point the Fed would have lost its ability to alter rates—cash doesn't pay interest nor can it be penalized—and would no longer be capable of exercising monetary policy. This is called the zero-lower bound, and it terrifies central bankers.

Fedcoin has the potential to alleviate the zero lower bound problem. Here's how.

As Fedcoin adoption grows among the public, cash would steadily be withdrawn. And while it might not shrink to nothing—the public might still choose to use some cash—at least the Fed would have a good case for entirely canceling larger denominations like the $100 and $50.

Consider also that it would be possible for interest to be paid on each Fedcoin  (unlike bitcoin and cash), the rate to be determined by the Fed. And just as Fedcoin could earn positive interest, the Fed could also impose a negative rate penalty on Fedcoin. This would effectively solve the Fed's zero lower bound problem. After all, if the Fed wished to reduce the rate on reserves to -2 or -3% in order to deal with a crisis, and reserve owners began to bolt into Fedcoin so as to avoid the penalty, the Fed would be able to forestall this run by simultaneously reducing the interest rate on Fedcoin to -2 or -3%. Nor could reserve owners race into cash, with only low denomination and expensive-to-store $5s and $10s available.

So by implementing something like Fedcoin, the Fed could safely implement a negative interest rate monetary policy.

(Lastly, monetary policy nerds will notice that the displacement of non-interest yielding cash with interest-yielding Fedcoin is a tidy way to arrive at Milton Friedman's optimum quantity of money, or the Friedman rule.)

The big losers: banks

Fedcoin has the potential to tear down the private banking system. Interest yielding Fedcoin would be able to do everything a bank deposit could do and more, and all this at a fraction of the cost. As the public shifted out of private bank deposits and into Fedcoin, banks would have to sell off their loan portfolios, the entire banking industry shrinking into irrelevance.

One way to prevent this from happening would be for the Fed to make an explicit announcement that any bank could be free to create its own competing copy of Fedcoin, say WellsFargoCoin. Like the Fed, Wells Fargo would promise to offer two-way convertibility between its deposits/cash/Fedcoin and WellsFargoCoin at a rate of 1:1 to ensure that the price of its new ledger entries were well-anchored. The bank could then implement features to compete with Fedcoin such as higher interest rates or complimentary financial services. Even as Wells Fargo's deposit base steadily shrunk due to technological obsolescence, its base of WellsFargoCoin liabilities would rise in a compensatory manner.

The resulting lattice network of competing private bank crypto ledgers built on top of the Fedcoin ledger would work in a similar fashion to the current banking system. Wells Fargo would make loans in WellsFargoCoin and take deposits of FedCoin as well as competing bankcoins, say CitiCoin or BankofAmericaCoin. Intra-bank cryptocoin payments would be cleared on the books of the Federal Reserve with reserves transfers over the Fedwire funds system, although Fedcoin might eventually take the place of Fedwire. A change in the value of Fedcoin or reserves due to a shift in monetary policy would be transmitted immediately into a change in the value of all private bankcoins by virtue of  the convertibility of the latter into the former.

Nor would it be necessary to start with Fedcoin and then introduce bankcoins. Why not begin with the latter and skip Fedcoin altogether? Why aren't private banks at this very moment switching out deposits and replacing them with cryptoledgers?

KYC: Know your customer

'Know your customer' regulations would make implementation difficult, but not impossible.

With bitcoin, the location of a coin (its address) is public but the identity of the owner is not. However, laws require banks to gather information on their customers to protect against money laundering. As these laws are unlikely to change with the advent of new technology, banks would probably require anyone wanting to use bank cryptoledgers to have an account with a regulated bank. This would not be too onerous given that most Americans already have bank accounts.  However, it compromises anonymity, one of the key ideals of bitcoin, since each coin would be traceable by the authorities to a real person.

Perhaps there is still a way to preserve some degree of anonymity. Historically the Fed has always been spared from KYC rules since it has never had to document who uses cash. By grandfathering KYC exemption to Fedcoin, any user who wanted to preserve their anonymity could use Fedcoin rather than any of the multiple bankcoin ledgers, just like today they prefer to use anonymous Fed cash rather than bank accounts to transact.

In summary

So that's a rough sketch of Fedcoin—a decentralized, flexible, and well-backed payments system that grants one user, the Fed, a set of special privileges and responsibilities. Feel free to modify the idea in the comments section.

And just so we are keeping tabs, these are the institutions that Fedcoin could eventually make obsolete: bank deposits, banks (unless the latter are allowed to innovate their own bankcoins), the credit card networks Visa and Mastercard, bank notes, Fedwire, and even bitcoin itself, which would be unable to compete with a stable-value copy of itself.

Bitcoin true believers may not like this post, but perhaps they can take something constructive from it. Fedcoin is one of the potential competitors in the distant horizon. Now is the time for the rebels to figure out how to create a stable-price version of bitcoin, before Darth Vader does it himself. Otherwise they may someday find themselves closing down their bitcoin startups in order to write code for the Empire.

Note: My apologies to readers for my having succumbed to the constant temptation to adorn all blog posts with Star Wars references.

Sunday, October 12, 2014

The market monetarist smell test

I gave myself a quick whiff this week to determine if I pass the market monetarist smell test. This is by no means definitive, nor is this an officially administered MM® test.

To be clear, my preferred policy end point is market choice in centralized banking. In other words, you, me, and my grandma should be able to start up a central bank. But that's a post for another day. First-best option aside, here's my reading of a few market monetarist ideas.

Target the forecast

**** 5 stars

Big fan. Targeting the forecast would take away the ad hoccery and mystique that surrounds central banks. We want central bankers to be passive managers of yawn-inducing utilities, not all-stars who make front covers of magazines.

First, have the central bank set a clear target x. This is the number that the central bank is mandated to hit in the course of manipulating its various levers, buttons, and pulleys. Modern central banks sorta set targets—they reserve the right to be flexible. Bu this isn't good enough. To target the forecast, you need a really clear signal, not something vague.

Next, have the central bank create a market that bets on x. Either that, or have it ride coattails on a market that already trades in x. If the market's forecast for x deviates from the central bank's target, the central bank needs to pull whatever levers and pulleys are necessary to drive the market forecast back to target.

The advantage of targeting the market forecast is that the tasks of information processing and decision making are outsourced to those better suited for the task: market participants. Gone would be whatever department at the central bank whose task is to fret over incoming data to determine if the bank is on an appropriate trajectory to hit x. Gone too would be the functionaries whose job it is to carefully wordsmith policy statements. The job of Fed-watching—the agonizing process of divining the truth of those policy statements—would disappear, just like lift operators and bowling alley pinsetters have all gone on to greener pastures. Things would be much simpler. If the market bets that the central bank is doing too little, its forecast will undershoot the target and the central bank will have to loosen. Vice versa if the market thinks the central bank is doing too much.

Targeting the forecast is the "market" in market monetarism. It's elegant, workable, and efficient—let's do it.

NGDP targeting

*** 3 stars

Meh, why not?

If we're going to target the forecast, we need a number for the market to bet on. Using the same target that central banks currently use is tricky. Most central banks are dirty inflation targeters. They try to keep the rate of change in consumer prices on target, but reserve the right to be flexible. Central banks have been willing to tolerate a little more inflation than their official target, especially if in doing so they believe that they can add some juice to a slowing in the real economy. Alternatively, they may choose to undershoot their inflation target for a while if they want to put a break on excessively strong output growth.

An NGDP target may be a good enough approximation of a flexible inflation target. NGDP is real GDP multiplied by the price level. If a target of, say, 4% NGDP growth is chosen, and the real economy is growing at 3%, then the central bank will only need to create 1% inflation. But if output is stagnating at 0.5%, then it will create 3.5% inflation.

So NGDP targeting affords the same sort of flexible tradeoff between the price level and real output that dirty inflation targeting affords, while serving as a precise number for markets to bet on.

The quantity of base money

* 1 star

Market monetarists have a fixation on the quantity of base money. This is where the monetarism in market monetarism comes from. Specifically, market monetarists seem to think that a central bank's policy instrument is, or should be, the quantity of base money. The policy instrument is the lever that the Carneys and Draghis and Yellens of the world manipulate to get the market to adjust the economy's price level.

But modern central banks almost all pay interest on central bank deposits. The quantity of money has effectively ceased to be a key policy instrument. (The Fed was late, making the switch in 2008). Shifting the interest rate channel (the gap between the interest rate that the central bank pays on deposits versus the rate that it extracts on loans) either higher or lower has become the main way to get prices to adjust.

This doesn't mean that the base isn't important. Rather, the return on the base is the central bank's policy instrument—it always has been. This is a big umbrella way of thinking about the policy instrument, since the return incorporates both the interest rate paid on deposits and the quantity of money as subcomponents. Reducing the return creates inflation, increasing it creates deflation.

Market monetarists seem to think that the interest rate channel ceases to be a good lever once interest rates are at 0%. But this isn't the case. It's very easy for central banks to reduce the return on deposits by imposing deposit rates to -0.5% or -1.0%. Going lower, say to -3%, poses some problems since everyone will try to immediately convert negative yielding central bank deposits into 0% cash. But if a central bank imposes a deposit fee on cash, a plan Miles Kimball describes more explicitly here, or withdraws high face value notes so that only ungainly low value notes remains, which I discuss here, there's no reason it can't drop rates much further than that.

If anything, it's the contribution of quantities to the base's total return that eventually goes mute. In manipulating the quantity of central bank deposits, central banks force investors to adjust the marginal value of the non-pecuniary component of the next deposit. Think of this non-pecuniary component as package of liquidity benefits that imbue a deposit with a narrow premium in and above its fundamental value. Increasing the quantity of central bank deposits results in a shrinking of this premium, thereby pushing their value lower and prices higher, while decreasing the quantity of deposits achieves the opposite. At the extreme, the quantity of deposits can be increased to the point at which the marginal liquidity value hits zero and the premium disappears, at which point further issuance of central bank deposits has no effect on prices. Deposits have hit rock bottom fundamental value.

So in sum: yes to targeting the forecast, and I suppose that an NGDP target seems like a good enough way to achieve the latter, and to hit it let's just keep using rates, not quantities. Does this make me a market monetarist?

Of course there's more to market monetarism than that, not all of which I claim to understand, but this post is already too long. Nor am I wedded to my views—feel free to convince me that I'm deranged in the comments.

Incidentally, if you haven't heard, Scott Sumner is trying to launch an NGDP prediction market.

Saturday, October 4, 2014

Stock as a medium of exchange

American Depository Receipt (ADR) for Sony Corp

You've heard the story before. It goes something like this. There's one unique good in this world that serves as a universal vehicle by which we conduct every one of our economic transactions. We call this good "money". Quarrels often start over what items get lumped together as money, but paper currency and deposits usually make the grade. If we want to convert the things that we've produced into desirable consumption goods (or long-term savings vehicles like stocks), we need to pass through this intervening "money" medium to get there.

This of course is fiction—there never has been an item that served as a universal medium of exchange. Rather, all valuable things serve to some degree or other as a medium of exchange; or, put differently, everything is money. What follows are several examples illustrating this idea. Rather than using currency/deposits as the intervening medium to get to their desired final resting point, the people in these examples are using a non-standard intervening media—specifically, listed equities—in order to move from an undesired currency to a preferred currency.

Zimbabwe and Old Mutual

In the midst of the Zimbabwe hyperinflation I began to toy with the idea of purchasing Zimbabwean stocks. The market value of the entire Zimbabwe Stock Exchange had collapsed to a fraction of its Zambian and Botswanan peers, and picking up a few bellwether names might provide some value, went my thinking. The difficult part was buying the Zimbabwean dollars necessary to build my position. Selling my Bank of Montreal deposits (I live in Canada) for deposits at a bank in Zimbabwe, say Barclay's Bank Zimbabwe, would not only take a long time to complete, but I'd end up having to pay the official rate for Zimbabwe dollars, which was far below the market rate. The losses on this forex conversion would destroy any opportunity for a profit on the shares.

There was an alternative route. I could sell my Royal Bank deposits for shares in a firm called Old Mutual, listed on the London Stock Exchange. The kicker is that Old Mutual had (and continues to have) a listing on the Zimbabwe Stock Exchange. Using a stockbroker in Zimbabwe I could have transferred my London-listed shares to the Zimbabwe Stock Exchange, sold the shares, leaving Zimbabwe dollars in my brokerage account. Since the ratio of Old Mutual in London and Zimbabwe was free to fluctuate (unlike the official exchange rate), I'd effectively be purchasing Zimbabwe dollars at the correct market rate, not the overvalued official rate.

Next I could have used my Zim dollars to buy the Zimbabwe-listed stocks that I wanted. When the time came to get out, I could have sold my shares for Zimbabwe dollars, repurchased Zimbabwe-listed Old Mutual shares, had my broker 'uplift' those shares over to the London stock exchange upon which I would once again sell Old Mutual for British pounds, eventually ending up with my Bank of Montreal deposits.

What an incredible chain of transactions! Which explains in part why I chickened out. Nevertheless, the example illustrates the necessity of having an appropriate and non-standard "medium of exchange", in this case Old Mutual, in order to shift from one brand of "money" deposits to another to another.

ADRs and the Argentinean Corralito

A similar example played out during the Argentinean corralito of late 2001 and early 2002. In early December 2001, in an effort to prevent massive capital outflows, Argentinean authorities established financial controls which, among other restrictions, imposed a ceiling of $1,000 a month on bank withdrawals. This became known as the corralito, the diminutive of corral, or animal pen. With a devaluation imminent, and even worse, pesofication—the forced conversion of bank deposits from USD into pesos—Argentineans could only sit helplessly as their frozen deposits awaited their doom.

Argentineans quickly found a way to evade the corralito. While they could only withdraw limited amounts of dollars from their bank accounts, they were allowed to buy any amount of stocks listed on the Buenos Aires stock exchange. Since stocks would be protected from the ensuing devaluation and pesofication, a mad rush into the markets ensued along with a terrific rise in share prices.

Snipped from Auguste et al, 2005.

What is interesting is that certain Buenos Aires-listed stocks were adopted as a convenient medium for escaping Argentina altogether. Here's how. An asset class called the American Depository Receipt, or ADR, trades on the New York Stock Exchange. ADRs are market-listed securities that represent an underlying batch of non-US shares. The way an ADR works is that a U.S. custodian bank will issue an ADR to an investor after the underlying shares having been deposited in a foreign depository bank where they will be held for safekeeping. An owner of ADRs enjoys all the economic rights (dividends, votes, capital appreciation) as the underlying shares held in deposit.

A number of Argentinean names traded on ADR form in New York, including a Banco Frances ADR and a Telecom Argentina ADR. During the corralito, an Argentinean could buy an Argentinean stock that traded on the Buenos Aires Stock Exchange, say Telecom Argentina, and immediately deposit these shares with a local depository. The shares having been deposited, a U.S. custodian bank would create an overlying ADR for the Argentinean investor. Since these ADRs were traded in New York, the Argentinean could turn around and sell the ADR for U.S. dollar deposits. Telecom Argentina shares and its linked ADR had become a medium of exchange of sorts, allowing Argentinean investors to convert from one brand of money, pesos, into another, US dollars.

Canada: Norbert's Gambit

Nor is the use of equity as a medium of exchange solely a phenomenon of crisis economies like Zimbabwe and Argentina. Enter Norbert's gambit. The name comes from Norbert Schlenker, an investment advisor in B.C. who popularized the technique. Canadian discount brokerages charge around 1.5% on forex conversions, which is a lot. Norbert's Gambit is a cheaper way to convert Canadian dollars to U.S. dollars and back.

The gambit works this way. Horizons US Dollar Currency ETF, which holds very short term US debt, trades on the Toronto Stock Exchange in US dollars under the ticker DLR.U as well as the ticker DLR, which is quoted in Canadian dollars. Investors can spend Canadian dollars in their brokerage account to buy DLR, convert those units to DLR.U, and then sell those DLR.U units for US dollar deposits. Voila, they've used an ETF as a medium to move from one "money" to another.

Interlisted stocks like Royal Bank or Potash Corp, which trade on both the Toronto and New York markets, can also be mobilized for Norbert's Gambit.


Norbert's Gambit, the Old Mutual switch, and the Argentinean ADR evasion are only a few examples of things that we normally don't consider to be "money" being mobilized as media of exchange. But these three are just the tip of the iceberg. Consider the fact that everyone who acts as a dealer in goods or securities is using these items as an exchange medium. Just as someone builds up a stock of cash for eventual future exchange, a t-shirt dealer purchases an inventory of t-shirts for future sale.

It's not just t-shirts. Every dealer from the gas utility to the car lot owner to a market maker in a small cap stocks uses the particular good in which they specialize—natural gas, cars, and penny stocks—as their medium of exchange. Some media are more general than others and will tend to appear in a larger proportion of transactions, but this doesn't make them qualitatively different from those that are less general. Put differently, there is no such thing as a valuable good that does not function as a medium of exchange: rather, there are only good media of exchange or bad ones.


There is a body of academic work on the Argentinean corralito, stock prices, and ADRs

1. Melvin, 2002., A Stock Market Boom During a Financial Crisis: ADRs and Capital Outflows in Argentina
2. Yeyati, Schmukler, & Van Horen, 2003. The price of inconvertible deposits, the stock market boom during the Argentine crisis.
3. Auguste et al, 2005. Cross-Border Trading as a Mechanism for Implicit Capital Flight: ADRs and the Argentine Crisis.
4. Brechner, 2005. Capital Restrictions as an Explanation of Stock Price Distortions During Argentine Financial Collapse: December 2001 – March 2002.
5. Lam, 2011. New Evidence on the Wealth Transfer during the Argentine Crisis.

No work has been done on Old Mutual and the Zimbabwean hyperinflation.

Sunday, September 28, 2014

The law of reflux

One of the coining press rooms in the Tower of London, c.1809 [link]

[This is a guest post by Mike Sproul.]

The law of reflux thus assures the impossibility of inflation produced by overexpansion of bank credit. (Blaug, 1978, p. 202.).

It is the reflux that is the great regulating principle of the internal currency; and it was by the preservation of the reflux, throughout all the perils and temptations of the period of the restriction, that the monetary system of these kingdoms was saved from the utter wreck and degradation which overwhelmed every paper-issuing state on the Continent… (Fullarton, 1845, p. 68.)

If you want to understand the law of reflux (and you should), then think of silver spoons. The silversmith shown in figure 1 can stamp 1 oz. of silver into a spoon. If the world needs more spoons, then silversmiths will find it profitable to stamp silver into spoons. If the world has too many spoons, then people will find it profitable to melt silver spoons. Unwanted spoons will “reflux” back to bullion. In this way, the law of reflux assures that the world always has the right amount of silver spoons. We could hardly ask for a simpler illustration of the Invisible Hand at work. But it costs something to stamp silver and to melt it, so the price of spoons will range within a certain band. It might happen, for example, that silversmiths only find it profitable to produce spoons once their price rises above 1.03 oz., while people only find it profitable to melt spoons once their price falls below 0.98 oz. If the costs of minting and melting were zero, then a spoon would always be worth 1 oz.

This is a point worth emphasizing: The value of a spoon is equal to its silver content. An increase in the demand for spoons would not raise the price of spoons much above 1 oz, since new spoons would be produced as soon as the price rose above 1 oz. A drop in the demand for spoons would not push the price much below 1 oz, since spoons would be melted when the price fell below 1 oz. Likewise, if the quantity of spoons supplied became too large or too small, market forces would restore the quantity of spoons to the right level, while keeping the price at or near 1 oz.

In figure 2, the silversmith starts being called a mint, and instead of stamping silver into spoons, the mint stamps silver into 1 oz. coins. The law of reflux works the same for coins as for spoons, always assuring that the world has the right amount of coins, and that the value of each coin always stays at 1 oz., or at least within a narrow band around 1 oz.

The usual thought experiments of monetary theory don't work in this situation. For example, economists often imagine that if the money supply were to increase by 10%, then the value of money would fall by about 10%. But the law of reflux won't allow this to happen. Mints would only issue 10% more coins if the public wanted coins badly enough to part with an equal amount of their silver bullion. And even if mints went against their nature and issued more coins than the public wanted, those coins would be melted or stored, and the value of a coin would not deviate very far from its silver content of 1 oz.

In figure 3, the mint re-invents itself again, this time as a bank. Rather than stamping customers' silver into coins, the bank stores the silver in a vault, and issues a paper receipt called a bank note. (Checkable deposits would also work.) This system has several advantages over coins. (1) It saves the cost of minting and melting. (2) It avoids wear of the coins. (3) Bank notes are harder to counterfeit, easier to carry, and easier to recognize than coins.

The law of reflux still works the same for bank notes as it did for coins. If the economy is booming and people need more bank notes, then people will deposit silver into their banks and the banks will issue new bank notes. If the economy slows and people need fewer bank notes, then people will return their unwanted notes to the banks and withdraw their silver.

Once again the thought experiment of imagining a 10% increase in the quantity of bank notes is pointless. Banks would only issue 10% more notes if the public wanted those notes badly enough to bring in 10% more silver. And even if we imagine that the banks took the initiative, printing 10% more notes and using those notes to buy 10% more silver, every bank note is still backed by 1 oz. and redeemable into 1 oz of silver at the bank, so every bank note remains worth 1 oz.

In figure 4, the bank makes one more important change. Rather than requiring customers to bring in 1 oz of actual silver to get a bank note, the bank also accepts an equal or greater value of bonds, bills, real estate deeds, or anything else that can fit in the vault. This system has several advantages: (1) Handling bonds, etc. is easier and safer than handling silver. (2) The silver formerly deposited can be put to productive use. (3) The quantity of bank notes is no longer constrained by the amount of silver available. (4) The bank earns interest on its bonds, bills, etc.

The law of reflux still operates as before, except that when people want more notes, they can bring in either silver or bonds, and when people have excess notes, the notes can be returned to the bank for either silver or bonds. As before, it makes no sense to ask questions like “What if the bank issues 10% more bank notes?” And as before, so long as every bank note is backed by, and convertible into, 1 oz. worth of assets, every bank note will be worth 1 oz. Just as reflux assures that the value of a spoon is equal to its silver content, reflux also assures that the value of a bank note is equal to the value of the assets backing it.

The Channels of Reflux

Here is a list of some of the many channels through which bank notes might reflux to the issuing bank:
1. The silver channel: Unwanted notes are returned to the bank for 1 oz. of silver. Alternatively, the bank sells its silver for its own notes, which are retired.
2. The bond channel: The bank sells its bonds in exchange for its notes, which are retired.
3. The loan channel: The bank's borrowers repay loans with the bank's own notes.
4. The real estate channel: The bank sells its real estate holdings for its own notes.
5. The rental channel: The bank owns rental properties, and tenants pay their rent in the bank's notes.
6. The furniture channel: The bank sells its used furniture for its own notes.

As long as enough reflux channels are open, it does not matter if a few channels are closed. Customers would not care if the furniture channel was closed, as long as major channels, like the bond channel, stayed open. The bank could take a more drastic step and close the silver channel, or could delay silver payments by 20 years, and as long as enough other channels stayed open, the law of reflux could operate as always, except that notes might be redeemed for 1 oz. worth of bonds, rather than 1 oz. of actual silver. The bank could even un-peg its notes from silver. Rather than redeeming a refluxing dollar note for 1 oz. worth of bonds, it could redeem dollar notes for 1 dollar's worth of bonds. As long as the bank's assets are worth so many oz., it doesn't matter if those assets are denominated in oz. or in dollars.

Once metallic convertibility is suspended, it would be an understandable mistake if people forgot all about the other channels of reflux, and started to think that bank notes were no longer backed by, or convertible into, anything at all. Unfortunately, this mistake has made it into the textbooks:
You cannot convert a Federal Reserve Note into gold, silver, or anything else. The truth is that a Federal Reserve Note has no inherent value other than its value as money, as a medium of exchange. (Tresch, 1994, p. 996.)
There you have it. The closing of just one channel of reflux (the metallic channel), has fooled economists into wrongly rejecting the idea that modern bank notes like the US paper dollar are backed and convertible. Once economists reject this simple and obvious explanation for why modern paper money has value, they are forced to resort to the more exotic explanations offered by textbook monetary theories, which are anything but simple and obvious.

Blaug, Mark, Economic Theory In Retrospect, 3/e. Cambridge: Cambridge University Press, 1978
Fullarton, John, Regulation of Currencies of the Bank of England (second edition), 1845. Reprinted by Augustus M. Kelley, New York: 1969.
Tresch, Richard, Principles of Economics, St. Paul, Minnesota: West, 1994

Wednesday, September 24, 2014

A brief history of the Guinea

1685 Guinea with the bust of James II (link)

The guinea makes a fascinating story because its evolution reveals so many different monetary phenomena. It began its life in 1663 in the Kingdom of England as a mere coin, one medium of exchange in a whole sea of competing exchange media that included crowns, bobs, halfpennies, farthings, not to mention all the foreign coins that circulated in England, Bank of England paper notes, as well as the full range of portable property—like jewelery and art—and property-not-so-portable, say houses and land and such. If things had stayed that way, the guinea's life would be a boring one and I wouldn't be writing about it.

But in the late 1600s the guinea crossed a line and became a very different thing. Rather than functioning as just one exchange medium among many, the guinea suddenly emerged as one of Britain's two media of account, the items used to define a nation's unit of account, in this case the £. Within a few decades it had wrested the medium of account function for itself, holding this pre-eminent spot until 1816, at which point the guinea was decommissioned.

Interestingly, while the guinea ceased to exist in 1816, its memory was sufficiently strong that it continued to function as a unit of account, albeit a relatively unimportant one, well into the 1900s. More on that later.

Just a regular coin

Whereas most of England's coinage at the time was silver,  the guinea was a gold coin. Introduced in 1663 during the reign of Charles II, it was initially rated at 20 shillings, or one pound (£), by the monetary authorities (the mint and the king). Pounds, shillings, and pence, or £sd, comprised the English unit of account—the set of signs that merchants affixed to their wares to indicate prices. The pound unit had been defined in terms of silver coins for centuries, but the the decision by the mint to give a 1 pound (or 20 shilling) rating to the guinea meant that the pound would now be dually defined in terms of both gold and silver coins.

However, according to Lord Liverpool, both the public and the authorities ignored this 20 shilling rating so that a market-determined price emerged for the guinea. In this way the guinea was no different from any other item of merchandise; its price floated independently according to the whims of buyers and sellers.

This stands in contrast to England's silver coinage. Silver pennies, halfpennies, and farthings had an extra function; they served as the nation's medium of account. The pound unit, the £, the symbol with which merchants set prices or denominated debts, was defined by the nation's silver coinage. Put differently, by setting a farm's price at £10, a seller was stipulating that the farm was worth the amount of silver residing in a collection of pennies and farthings.

When something serves as the medium of account, it's price doesn't float independently. Rather, the whole universe of other prices shifts to accommodate changes in the value of the medium of account. For example, if the value of silver were to have risen in the 1670s due to increased demand for silver jewelery, then the entire English price level would have had to fall. Alternatively, if the amount of silver in the nation's coinage was debauched, then the English price level would have risen. A change in the demand for gold in the 1670s, however, would have produced an entirely different result; the relative price of the guinea would have shifted, but little else. That's why a medium of account is so special. Unlike all other items, the price of everything pivots around it.

The fact that the guinea's initial 1663 rating had been ignored was very important. Imagine that the authorities had been stern about enforcing it. Returning to our farm example, in setting the farm's price at £10, our seller would have been stipulating that the farm was worth either the amount of sliver residing in a collection of pennies and farthings, or the amount of gold residing in the guinea. A very different monetary system would have emerged; bimetallism. But more on that later.

Liverpool tells us that the guinea fluctuated between 21 and 22 shillings in its first decades, but in 1695 its price rose rapidly to 30 shillings. This wasn't because of an increase in the demand for gold but a function of the quickening pace of clipping and sweating of pennies, which reduced the quantity of silver in the coinage. Guineas weren't the only commodity to rise in 1695; the entire array of English prices had to pivot around the diminishing value of the silver penny. Once the silver coinage was reformed (its silver content being restored) in the Great Recoinage of 1696, the price of guineas quickly returned to 22 shillings.

The switch to bimetallism

Things all changed in 1697 when the Exchequer, the department responsible for receiving taxes, announced that all guineas were to be accepted by the Exchequer's tellers at 22 shillings. Prior to then, the Exchequer had accepted guineas at the going market rate. As Sykes points out, after 36 years of floating this was tantamount to fixing the price of the guinea relative to silver. Guinea couldn't circulate for less than this stipulated amount, say 21 shillings, because an arbitrageur would mop up those guineas at 21 shillings and use them to pay 22 shillings worth of taxes, earning him or herself a 1 shilling gain. (The next year, the Exchequer would reduce this rate to 21 shillings 6 pence.)

Britain, which had been on a silver standard up to 1697, was now on a bimetallic standard, with the £ unit defined as the amount of silver residing in the English penny, and simultaneously the amount of gold residing in the guinea.

The guinea takes over

The problem with the new standard was that in setting the guinea at 21s 6p, the Exchequer had overvalued gold relative to the market price, more specifically the silver-to-gold ratio prevalent in the rest of the world. By how much? In 1702 Sir Isaac Newton, Master of the Mint since 1699, concluded that 'Gold is therefore at too high a rate in England by about 10 pence or 12 pence in the Guinea.' In other words, the Exchequer should have announced it would only accept guineas at around 20s 6p, or 4.6% less than it had.

What were the consequences of this over-valuation? All of the silver pennies began to leave Britain, gold coins filling the void. Given the choice between paying a debt or a tax in either an overvalued or undervalued instrument, people will always select to use the overvalued one. After all, buying 20 shillings 6 pence's worth of gold in France and using it to discharge a 21s 6p shilling tax liability in England resulted in a 4.6% profit (less transportation and minting costs). The undervalued instrument, in this case silver, is best used in other parts of the world where it is capable of purchasing a larger real amount of goods (or discharging a larger real quantity of taxes) than in the country in which it is artificially undervalued. This is, of course, Gresham's law; the bad drives out the good.

So our guinea, which had started its young life as a mere medium of exchange, had not only graduated to becoming one of only two English media of account, but was responsible for the mass flushing out of silver from England.

By 1717, the silver outflow was getting significantly bad that the authorities decided to do something about it. Newton, still Master of the Mint, noted that the market price for the guinea was around 20s 8d, given the exchange rate between silver and gold in other European markets, and suggested an initial rate reduction from 21s 6d to 21 shillings.

But even at this lower price the English authorities were still overvaluing the yellow metal. They had now fixed the silver to gold ratio at 15.069 to 1, but because the rate was 14.8 to 1 in Holland and France, a profit still remained on exporting silver and importing gold. This silver outflow would continue over the decades until most silver was gone. England had gone from a bimetallic standard to a monometallic standard. Though it was still de jure bimetallic, de facto it had become a gold standard. And the guinea, which was now the controlling element in English prices, was to blame (or at least the decision to misprice it was).(1)

The end of the guinea

For the next century, the English price level pivoted around the value of the gold guinea, until the Great Recoinage of 1816, at which point the guinea's life suddenly came to an end. Since the days of Isaac Newton, the guinea had been awkwardly rated at 21 shillings, or one pound one shilling. This must have made payments somewhat arduous since there was no coin that could satisfy an even 1 pound bill or debt, and people like round numbers. The decision was made to introduce a less awkward gold coin, the sovereign, with slightly less gold. The sovereign was conveniently rated at exactly 1 pound, or 20 shillings, the upshot being that the pound unit of account still contained just as much physical gold as before, but now a coin existed that corresponded with the exact pound unit. The guinea was dead.


Well, not entirely. Though is was no longer being minted, the guinea continued to be used as a way to price items. According to Willem Buiter (pdf), auction houses and "expensive and pretentious shops" continued to set prices in terms of guineas through the 1800s and 1900s. Bespoke tailoring and furniture, for instance, was quoted in the legacy gold coin. The unit used was g, or gn, with the plural being gs or gns, although payments were made in sovereign coins or Bank of England notes.

Gillette advertisement (link)

Doctor's and lawyer's fees, often known as "Guinea fees" we're advertised in terms of the legacy gold coin. Whereas common laborers were paid in pounds, payments in guineas was considered more gentlemanly. You can see it pop up in the literature of the time. In Arthur Conan Doyle's Sherlock Holmes tale the Adventure of the Engineer's Thumb a stranger offers Mr Hatherly, a hydraulic engineer who is down on his luck, a unique proposal. "How would fifty guineas for a night's work suit you?"

An ad from 1929 (link)

The standard rate paid by Charles Dickens for contributions to his weekly periodicals Household Words and All The Year Round was half a guinea a column or a guinea a page. In his novels, the guinea pops up often. In Oliver Twist (set in the 1840s), a 5 guinea reward for information on Oliver is posted by the kind Mr. Brownlow.

In more modern times, horses continue to be auctioned in terms of guineas.

Dancing Rain sold for 4 million guineas (link)

Now of course this is a bit of a come-down for the once almighty guinea. Serving as the unit at Tattersalls isn't the same as underpinning the entire price level. But at least its better than the sovereign, the coin that replaced the guinea, which has gone silent, or most other medieval coins for that matter, which neither circulate nor serve as legacy units.

(1) The 1717 reduction of the guinea to 21 shillings was accompanied by the requirement that those guineas be accepted as legal tender at that price. Prior to then, only silver had functioned as legal tender, meaning that a debtor could only discharge a debt with silver coins. After the change, a debtor could choose to use either guineas or silver coins to pay off their debt, a decision made easier given gold's overvaluation.

Some References:

Lord Liverpool, A Treatise on the Coin of the Realm, 1805.
Sargent & Velde, The Big Problem Of Small Change, 2001.
Selgin, Good Money, 2008.
Sykes, Banking and Currency, 1905. 
Macleod, Bimetallism, 1894.