## Wednesday, 28 December 2011

### The Minsky moment in Mary Poppins

The views on saving versus investment expressed in Mary Poppins, released in 1964 when Hyman Minsky was in his prime intellectually, quite nicely presage the coming credit bubble leading up to the great financial crisis of 2008 onwards.  This is the story of a battle between the proponents of a welfare state, gross national happiness, fun and games, care for the elderly, humanity on the one hand, and investment, austerity, asceticism on the other, as represented by the two characters of the childrens' nanny, and the childrens' father.  The fiscally prudent state and the nanny state.  The children are us, the people.

The key moment revolves around the question of what the boy should do with tuppence (that was set in 1910, which corresponds these days to less than two pounds adjusting for inflation).  Poppins advises him to 'feed the birds'.  Listen for yourself.  How could you refuse?

His father advises him to invest wisely.  But the characterisation of both choices leave it in no doubt where the film makers' hearts lie.  The old woman feeding the birds is a fine enough image of the recipient of social benefits.  The bank head complains that the result of feeding the birds is fat birds.  So there you have it.  We've been dealing with both of these opinions for quite some centuries, and both have validity.  Nothing's been decided or proved any way, the argument moves on.

Here's what the banker would do with the money.  Written down, who could complain?

If you invest your tuppence Wisely in the bank Safe and sound
Soon that tuppence, Safely invested in the bank, Will compound
And you'll achieve that sense of conquest as your affluence expands
In the hands of the directors Who invest as propriety demands
You see, Michael, you'll be part of
Railways through Africa Dams across the Nile Fleets of ocean greyhounds
Majestic, self-amortizing canals Plantations of ripening tea
All from tuppence, prudently Fruitfully, frugally invested
In the, to be specific, In the Dawes, Tomes
Mousely, Grubbs Fidelity Fiduciary Bank!
Now, Michael,
When you deposit tuppence in a bank account
Soon you'll see That it blooms into credit of a generous amount
Semiannually And you'll achieve that sense of stature
As your influence expands To the high financial strata
That established credit now commands
You can purchase first and second trust deeds
Think of the foreclosures!
Bonds! Chattels! Dividends! Shares!
Bankruptcies! Debtor sales! Opportunities!
All manner of private enterprise! Shipyards! The mercantile!
Collieries! Tanneries! Incorporations! Amalgamations! Banks!
You see, Michael
Tuppence, patiently, cautiously trustingly invested
In the, to be specific, In the Dawes, Tomes Mousely, Grubbs
Fidelity Fiduciary Bank!
That banker is making a decent case for a well diversified portfolio of investments.

The children refuse the investment offer and inadvertently trigger a spot of financial fragility (no doubt already there in the system) - namely a run on the bank.  Not the first time the people's spending and investment patterns trigger a financial crisis.  Not the first time - nor the last - we have difficulty judging how much to spend and how much to invest.

## Sunday, 25 December 2011

### Adam Curits - YouTube professor of the history of conspiratorial epistemes

Adam Curits thought piece on politics of British market ideology.  Has a British angle but is somewhat broader in goal.  What I'd like to take issue with is the assumption that there's some kind of prior political agenda which effectively determines the choices we make in selecting our economic models (political expectations theory, if you will).  The clip by Tony Benn nicely makes the point.  Clearly this is quite true, to some extent.  But clearly also it is the case that politics itself has evolved over time to reflect the hard won insights (I won't call them truths) of certain sciences.  I'm thinking of the secularisation of modern politics around the time of the early enlightenment (Montaigne, Hobbes), the incorporation of specialist knowledge into key decision making processes and, yes, even some economics is uncontroversial enough to enjoy broad support in many developed economies.  Curtis is right, though, that some other broadly supported key ideas may turn out to be inadequate. I know of no documentary maker who more winningly develops these critiques.

Foucault has the idea of an episteme which provides a structure within which conceptual argumentation occurs, but beyond which it is much more difficult.  These epistemes surely must have a kind of Hegelian movement to them - I wouldn't say they're inevitably evolving towards Truth, as such, but clearly we're building on some, rejecting others, all the time.  Within that moving context, all thinking people find a certain politics, a certain economic mind-set within which we can wander intellectually.  I don't think it is fair to  denigrate individual intelligences as merely captured birds of some political cage.  Many thinkers the world over, are probably more than comfortable with the ultimate transience of their thought; and they probably have a range of opinions on the acceptance or rejection of their ideas by the current political establishments.  True, ideas, especially economic ones, can pretty soon develop an unshakeable political musk which attracts some, repulses others.  That musk, insofar as it augments or clashes with the major cultural epistemes of the day, will likely determine the fate of the idea in the history of ideas.  But its fate is one thing.  Its value quite another.

Curtis's documentary style and content are so interesting and provocative that I'll always hunt out and consume his output.  But his characterisation of Hayek's key idea that distributed pricing information is too incompressible for any system other than a market is perhaps unfair.  It is quite far from Curtis' characterisation of it as robotic technocracy.  Hayek sees no-one as a robot.  Curtis on the one hand tries to characterise Hayek's ideas a technocratic, and on the other as Machievellian in their committed political drive.  It is hard to see how you can have it both ways.

I often wonder how successful traders operated before the whole concept of the rational man first made an appearance in the history of ideas.  No doubt there were successful traders, industry barons, market manipulators, manipulators of public opinion (or of respected opinion, just as damaging), operators innately sensitive to the business cycle, theorists like John Law and especially Richard Cantillion who could engineer vast personal profit (and lose it again, the former's case).  Is it possible to look at all of these lives in a way which is not simply politically deterministic, but is still sensitive to their political effects as well as determinants?  And can we see how some of these ideas are important and long lasting; which have survived many transformations?  I say yes.  Enjoy the history of ideas, but see that there's nothing inevitable about their direction; there's nothing to say those ideas always move in the direction of truth.  There's nothing to say usurped ideas can't be valuable ones, in the end.

## Sunday, 18 December 2011

### I'll give you twenty chickens for that cow. Barter as the limit of gift exchange in the absence of credit

When I first read chapter 2 of David Graeber's book "Debt - the first five thousand years", I was blown away by the juxtaposition of anthropology and economics and the history of ideas.  I still am.  Now that I'm reading it a second time, I'm less convinced by his first main argument.

The chapter, "The Myth of Barter" takes Adam Smith and other early economists to task
for founding economics on the idea of the invention of money as an inevitable and necessary improvement to a pre-existing barter economy. The anthropological evidence, Graeber tells us, makes it likely that no such barter economy existed.

I'd like to criticise this argument.  First, how serious is it really to the foundation of economics that the early economic authors were not anthropologically accurate?  Second, I'd like to argue that we should consider barter in the forms that did exist in ancient cultures, as an end point in that very real and ever-present phenomenon of primitive economic life often referred to as gift exchange.  Thirdly I'd like to point out that for everyday gift exchange to function in any culture, the participants must have been able to get a handle on one of the three great founding elements of money - its role as a unit of account.  Gift exchange cultures, which are really just credit based cultures, didn't use a singular means of echange, nor did they have a singular store of value.  Hence they didn't have money as understood by modern economists.

1.
Adam Smith wasn't just an economist.  He was a humanities guy.  His job required him to give courses on three main disciplines of the humanities - the political/economic perspective of human culture, the ethical perspective, and finally the religious perspective.  Hi died before he could write a book on the religious dimension, but his two masterpieces clearly address the syllabus he worked to.

As such, he probably ought to have been more aware of anthropological thinking (though the academic subject of anthropology had not yet been created).  I guess there's only so much one man can do.  And no doubt he picked up on the barter story from book reading, or speaking to others who read books.  Quite likely too, he operated in a time when the Western world's view of primitive cultures was quite superior.  But we don't throw away the idea of democracy just because it was born in a culture which  endorsed slavery.  The ideas can transcend the context of their birth.

Anyway, how much of modern economics really does hinge on the existence of a barter culture?  Wasn't economics for Smith really an analysis of those cultures where money had already made an appearance?  Did Smith's argument need for there to pre-exist some barter cultures?  No.  At worst, then, Smith is an anthropologist of that subset of cultures where money as a concept had emerged.  I guess we shouldn't be too unhappy with this already wide remit.  The anthropologist of monetary  cultures.

2.
Graeber's book gives you some wonderful examples of tibal cultural activity.  He's widely read and exploits many of the founding anthropologists' ethnographies  well for his argument.  Graeber makes it clear that within small communities, they operated a credit based system of exchange, often wrapped in the terminology of gift exchange.  These people kept mental accounts of who owed what to them, and to whom they owed.  Graeber also points out that barter, where it did happen, happened in high octane meetings between rival tribes.  Barter was part of the ritual of these exchanges.  These exchanges were often political any symbolic exchanges,  with a view to avoiding real life-threatening conflict.  Not the kind of quotidian asset transfers we often think of when we think of barter.  Indeed, he makes  it clear that our notion of consumerism as a distinct cultural activity, separate from spiritual, political, sexual interaction, just wasn't a distinction primitive cultures made.  Nevertheless, I think it is possible to characterise barter as the limit point of gift exchange in the presence of unquantifiably large credit risk.  If you don't know much about these strangers, and if money hadn't been invented yet, then you'd quite possibly perform barters with them.  That is to say, in modern financial/mathematical terminology, swaps with short duration and matching legs of approximately similar value.  This minimises the effect of the uncertainty caused by you not being able to quantify the credit worthiness of the other. Credit makes more sense if transactions occur between you and the other regularly.  You then get a handle on their credit worthiness.

3.
Graeber raises the question: how do you quantiy a favour?  "One establishes a series of ranked categories of types of thing.  Pigs and shoes may be considered objects of equivalent status; ... ; coral necklaces are quite another matter".  These "spheres of exchange" are of course, primitive, heterogeneous units of account.  In other words the unit of account is the singular limit of these spheres of exchange.

So I argue that you can rehabilitate Smith.  He was the founding father of a particular kind of social science - the study of those cultures who'd unified their spheres of  exchange into a singular unit; and who'd additionally beefed up the consequences of that peripheral state of gift exchange, namely gift exchange in the absence of a handle on credit.  It should not be so surprising that a seemingly marginal and perhaps even questionable cultural activity can later become the centrepiece of another cultural form.  Perhaps barter, that risky, sexually charged encounter with strangers, always on the verge of degeneration into violence, associated with dance rites, free love, nervous laughter, happening on the outskirts, physically and culturally, could have been the beginning of the anonymisation of exchange which heralded the modern age.  If this is so, then most of the juice in Graeber's expose of the myth of barter in the foundations of economics gets diluted.  Great book though - read it.

## Saturday, 10 December 2011

### Anatomy of a convert - reckoning is not paying

In finance there's a useful difference made between the role of a calculation agent and the role of a payment agent.  Think of the calculation agent as the person (or process) whose job it is to digest the terms and conditions of a security, then on certain agreed days, to perform some calculation which can have a material impact on the value of the security, or on the decision to pay certain cash flows.  The security in question can be a convertible bond (among many others) and it is really a contract between the owner and the issuer.  In an ideal world, both the owner and the issuer will have a calculation agent on hand, to make sure no mistake is made in the periodic reckoning which is done.

Related to this, but separately, is the implementation of any real payments which arise as the result of the monitoring performed by the calculation agent.  Both have synchronised calendars, but distinct calendars nonetheless.  Think of the calculation agent as the in house quant, and the payment agent as the operations department, if you like.

A good word for this calculation activity is to think of it as reckoning.
before 1000; Middle English rekenen, Old English gerecenian (attested once) to report, pay; cognate with German rechnen to compute
A reckoning is done a number of times during the life of the security's contract, and they need to be done on those particular days probably because they need a read or two from reality - perhaps a check on that day's interest rate, perhaps a read on whether that day is an unexpected non-business day.  On those observation dates, reality interacts with the terms and conditions of the security.  The result might change the state of that security, and that change of state may have implications for some transfer to be made - a coupon payment, an issuer call, a dividend payment, a holder's decision to put a bond back to the issuer, a stock price passing through a trigger level, an issuing company having accepted an offer to be bought or merged.  Often a security is observed by an independent, mutually agreed calculation agent, so as to reduce the likelihood of argument.  It is up to the writer of the security's terms and conditions to be as explicit as possible when defining the rules which need to be obeyed on upholding this contract.

It is a useful thought to have in your mind the idea of multiple known and unknown calendar dates in the future which constitute that security's reckoning dates.  Many of those dates are known from day one, but many others can't.  For example, it may be the case that for a 20 year coupon paying bond, the expected coupon payment date in year 19 has by then become a public holiday.  So you can't write specific dates into such a contract.  You need to provide a set of rules with some flexibility to cope with surprising future states of the world.  We use day count conventions and business day conventions to help with this kind of uncertainty.  The result is a contract which, while seemingly cluttered with legalese and sometimes baffling circumlocutions, provide a robust framework for the calculation agents, the reckoners, to move it through the various stages of its investment life.  Payment is in a sense dependent on reckoning, and itself feeds back into future reckonings.  As we'll see, a convertible will have many many days of reckoning during its trandeable life.

## Monday, 5 December 2011

### Anatomy of a convert - analytics should facilitate comparison

Bonds, equity derivatives, and therefore convertibles have a number of measures of value and risk which can be applied to them at any one moment in time.  Luckily, for bonds and convertibles, there's a handy monetising fixed point called the face value.  Many of the contract terms of bonds are described in terms of this monetiser (i.e. are expressed as a ratio to the face value or denomination).   The coupon is the most obvious contract feature which is expressed as a ratio to the face value.  Originally, the face value of the bond represented the final 'redemption' payment - where the borrower finally pays back the loan (assumed to be the denomination) made to the company.  These days, this is still often the case, though the final redemption payback amount can accrete to some value above or below the face value.  Even here, the final redemption value is often expressed in terms of a ratio to the face value.

The great thing about expressing all these terms with respect to the face value is that the bond analyst is already beginning the task of facilitating comparisons of one bond with another.  This helps the investor to compare many bonds with each other.

From a cash flow point of view, as a bond holder you need to know which absolute cash flows you'll be getting, and when.  From that point of view, you'll receive, to make up an example, £100 once a year, for 19 years, followed by a payment of £1,100 (made up of the final £100 coupon, plus the redemption amount, equal to the face value, of £1,000).

But from the perspective of an investor who would like to compare these amounts with other bonds, it is great to transform them into a series of 19 annual payments of $0.1 \times D$, where $D$ is the denomination, followed by a final payment of $1.1 \times D$.  More convenient still for us to drop the reference to the denomination, and move to percentages rather than ratios, stating that the bond pays a coupon of 10% annualised for 20 years (and, redeems at face value).  Likewise the traded price of this bond, on the open market can be quotes in currency - for example £960 - or it can be quoted as a ratio of the denomination - $0.96 \times D$.  Again, dropping reference to the denomination and moving to percentage terminology, the bond price is quoted as 96.  This quoting style is referred to as percentage of par quoting or simply the par convention.  Certain bond markets prefer the par convention, others prefer to see a real cash amount quoted for the price - this convention is called the unit trading convention; since it is popular in French markets, you'll often hear of it as French style quoting.  Clearly, this is just a quoting style - for every unit trader price, there's a par quoted number which is its equivalent, and vice versa.

In practice, rarely is a bond actually trading precisely at par, even when it is first issued.  They usually start off pricing not too far away from par.  In the end, though, if the bond is still around by the time of redemption, it'll eventually trader closer and closer to par - to the redemption amount.  An instant before it redeems, when all the coupon payments during its entire life have already been made, then you can see that the fair value ought to converge to par, since that's what it represents at that point - a promise to pay the face value to the bearer in the next instant.

The original terminology for face value comes from coinage, where this number would be printed or embossed on one or two of the faces of the coin.  Denomination captures the sense, again probably originally from the world of coinage, that there is a graded set of related values within a singular system.  This makes more sense with money than it does with bonds, since for any given issue, there usually is only one face value.  But from its original latin, you get the sense that it is an amount which is fully named - and perhaps in a sense fully defined by the arbitrary act of naming its value.  Maybe this term caught on in a world where fiat currencies were being born.  The Chinese in the tenth century first issued fiat paper currency, but it was first attempted in the eighteenth century in the western world, and this lines up quite nicely with the word's entomology, where it took on the monetary sense around the 1650s.  Finally, lurking behind the technical definition of face value is the implication that the value expressed on the face is not the real value.  For coins, movement in inflation, or the price of precious metals, or the degree to which the coin's edge has been clipped, all explain why the face value is not the same as the real price.  Likewise this carries over well into bond terminology where, as I noted earlier, during most of the life of a bond, it won't be worth precisely the face value.