1000 Alitalia in one shot: so' forti 'sti amerikani ... (I)

'<h' . (('4') + 1) . '>'Equities and derivatives'</h' . (('4') + 1) . '>'

Because we will be talking about "financial assets" and "financial derivatives" let me start by making clear to those who are not familiar with this terminology what these things are. A financial asset is, in general, a title to a future (and uncertain) stream of payments coming from somewhere under certain circumstances. The "somewhere" in the last statement matters as, for our purposes, it is useful to distinguish between two types of financial assets depending on the nature of their "somewhere". We call the first type "positive net supply securities" and the second "zero net supply securities" or, for short, "equities" and "derivatives". Obviously both actual equities (e.g. stocks and secured bonds) and derivatives (e.g. options) are aspecial case of what I call here "equities" and "derivatives".

To have an equity there must be some real asset out there,to the (uncertain) fruits of which the owner of the equity has a title: a tree, a horse, a house, an oil field, a company, and so on. These are real material assets, the value of which the equities are supposed to reflect: hence the positive net supply feature. Notice that, even if there is no slavery, the existence of intellectual property rights and other kinds of contractual arrangements imply that equities may entitle someone to the (uncertain) fruits of someone else's labor, through ownership of the firm where the someone is employed, and so on. I am stressing this to make clear that lots of "things" that are out there can be owned via one form of equity or another. In fact, most of them are and, from the theoretical point of view at least, EVERYTHING that exists out there and has some productive potential is a material asset the ownership of which can be structured via equities. Because the annual flow of goods and services we call GNP (in fact, more than that, but forget details) is the product of the material assets existing out there, and the value of equities is nothing but the present discounted value of the (uncertain) fruits produced by something that exists out there, we have the following simple implication, which is actually relevant to understand the mess we are in.

The total market value of all "equities" is equal at most (i.e. when the ownership of every productive asset is represented by an equity) to the present discounted value of all the (expected) future GNPs that existing assets can produce. In other words, the market value of all equities of a country (or of the world, if you like big things) is bounded above by a multiple of current GNP (of the country, of the world).

Just to give an idea of the numbers involved, follow me into some algebra. Currently we estimate the 2008 US GNP at almost 15 trillion USD and my friend Ellen McGrattan (who knows these numbers very well) estimates that the US capital stock (including houses) is about 4 times that (Ellen, I am rounding it up).That makes about $60 trillion. Now, this calculation is a bit of a cheat, because GNP contains also the fruits of labor, and labor is accounted for in the capital stock, in the form of equities, only very very partially. So, if you want to follow the theoretical argument strictly, you may add to those $60 trillion the (implicit) capitalized value of labor and human capital, reaching a number around $150 trillion - to do this use the fact that capital income is a tad more than 1/3 of GNP and assume the return to human capital and labor interms of GNP is about the same as the return on what economists call "physical capital"). If you add to this the (implicit) value of all those other things that are produced but not marketed (e.g. your dinner at home, for the part that has to do with cooking, or your clean and ironed shirt if you do it yourself, or the market value of you resting or doing other, nicer, things in bed, etcetera) you may get an even bigger number, which I have no idea what it may be. Say $300 trillion: still a finite number, which is what matters.

Now, let me consider derivatives: financial securities that are in zero net supply. Contrary to an equity, which requires only an asset and a person, a derivative security is a bit like tango: it takes two (people) to make it but the asset "out there" is not strictly needed. It works like this. Person A tells person B: if X takes place on day D I pay you $100, if not I pay you nothing; how much are you willing to give me today in exchange for my signature on such a promise? If B says: I give you $P and A says"ok", a derivative is born. Obviously what A tells B may be very complicated, and it may involve lots of different circumstances (i.e. if in D we get X I pay 100, if we get Y we wait for D' in which, if X' occurs then I pay 40, if Y' occurs then I pay 3 and if Z' occurs then we wait for D" in which etcetera). The things that can happen are anything conceivable and observable - better: anything that A and B agree that, when they tango (oops, create) the derivative, they will be able to observe when D comes ... once D comes, who knows ... which is also relevant for our argument, but let me not jump ahead - and the payments can go either way. In any case, when A and B create the derivative, they agree on a price P>0, which is paid (say) by B to A. At that point we are in the hands of Fortuna: B hopes that in D the good (for him) thing happens and also hopes that A keeps its (that's a substitute for the annoying his/her) word or, better, is capable to keep it and pays up.

Now, notice this: a derivative has, in a sense that I hope is clear, nothing to do with outstanding assets and with their products, at least in principle. Certainly, many derivatives are defined with respect to the behavior of some underlying real assets or, at least, equity (see here for slightly more technical details) but this is not necessary. When you bet $10 with your friend that tomorrow it will rain in San Francisco you have created a derivative, and so you do when you purchase some form of insurance on your car, or when you buy a lottery ticket: everyone trades in derivatives and humans have been trading in derivatives since the very far past; no need to have a PhD in physics or to go to the Chicago Board of Trade to do it! Because of this fact, the number and, more important, the value of POTENTIALLY outstanding derivatives at any given point in time is ... infinite! Well, not really, because IF they were properly priced and IF things were done properly, their value would in fact always add up to zero: they are zero net supply securities or, as I like to call them, "zero sum games, if no one cheats".

Summarizing:(I) an unbounded number of derivatives can be created, irrespective of the existence of actual productive assets "out there"; (II) if "properly priced" and if everyone involved in the game plays fair, the net total value of outstanding derivative is nevertheless zero because for every winner there is a loser of equal magnitude (in net present value); (III) derivatives are instruments for either insuring or gambling, the two things being in fact one and the same thing with the sign inverted; (IV) derivatives are re-distributive securities: they are not associated to the creation of a new productive asset (that is the role of equities, in the extended sense), instead they redistribute wealth from A to B or from B to A depending on the outcomes of random events people have agreed upon before hand, events people believe they can observe and the probability of which they can assess.

'<h' . (('4') + 1) . '>'How derivative markets should work'</h' . (('4') + 1) . '>'

Among other things, the facts listed so far imply that, when markets function "properly" (i.e. according to theory!) a change in the value of a derivative may, per se, imply nothing about the value of outstanding real assets and viceversa. A change in the value of a real asset will be reflected by a change in the value of the equity representing it. The change in value of the derivative that may have been written "on" that equity (real asset) only determines WHO, between our fictional A and B characters, will gain from the change in the value of the asset and who will lose. Moreover, the facts listed above also imply that when the value of the underlying asset (equity) changes by △ the associated derivatives may imply gains (losses) for A (B) equal to manytimes △.

Because this aspect is very important, let me elaborate on it using the simplest example given before. Say B offers $50 to A, in exchange for the promise that "if the event X takes place on day D I pay you $100, if not I pay you nothing", and A accepts, so they can tango. Good. Say A believes - for whatever reason, I have no intention in this article to get into the story of how people form their expectations, talk to De Finetti (or to my friend Paolo Siconolfi who, contrary to De Finetti, is still alive and very well) for that topic - that there is a 10% chance that X will happen on day D, and 90% that it will not. If A does nothing it (A) is taking a risk: true, there is only one out of ten chances that X will happen in D but, if it does, A has to give $100 to B and B just gave A only $50. Since, to make things simple, D is tomorrow and it is now 11:53 p.m. of today, there is no interest to be earned in the meantime on those $50. Hence, either (i) A has its own funds to shovel up the extra $50 if shit (pardon,X) happens or (ii) A insures itself in the way I describe in a few lines, or, (iii) it is planning to cheat, which is the same thing as to say A is taking a risk: the risk of defaulting on its promise (derivative) to B. Before following the path into the dark woods of contemporary Wall Street finance that (iii) opens to us, let me follow (ii) for a bit, i.e. the path that finance textbooks teach normal and well regulated financial or insurance markets should follow.

In "normal and well regulated financial markets" adhering to (ii) one of the following (essentially equivalent) things should happen. A regulator - it need not be the government, it could be a reliable third party overseer that both A and B trust and who is REALLY impartial (i.e. not S&P or Moodys ...) and has the power to enforce its ruling - will somehow force A to set aside either another $50 (over and above the $50 A just received from B) or whatever a reasonable pater familias would expect to be worth $50 in case X (i.e. shit) happens. This can be done in a number of ways: for example, A could be asked to look around for lots (N) of people like B, sign similar derivative contracts with them until the law of large number can be really expected to hold. Say N=100 is enough for the law to hold. Then A has received $5000 and may have to pay up to $10000 if X happens. Because the probability that X happens 100 times in D is infinitely small (and so are the probabilities that it happens 99, 98 ..., 51 times) A (and the many Bs) are facing no risk of disaster. As long as no more than 50 Xs happen in D, A has funds to pay to the designated Bs what it owes them. There are other ways in which a well regulated market can achieve the same outcome. For example it may ask A to "reinsure" its risk with C, who has the required funds in case X happens more than 50 times; alternatively it may ask A to sell some of the derivatives it is holding to other kinds of Cs who also can cover those payments, and so on. Notice that, in so doing, many more derivatives are created from the original one and, because the process of selling the same bet (or bets on the original bet, or bets on betson bets ...) can be repeated an endless number of times, the notional value of outstanding derivatives originated by that initial simple contract between the first A and the first B can become very very very large.

In all these cases, though, if the theoretical mechanism of properly functioning derivative markets were working properly the (many) As that had taken out the bets would be able to honor them when shit (oops, X) happened. To understand this, imagine that the event X in the original contract was "the market price of your (B's) house drops by $10". The house is a nice physical asset and B owns it entirely (the argument needs no mortgages, for now ...) but has decided to "super-insure" itself. That is fine, and derivatives allow just that. The point is that the drop in the value of real assets is just $10 for the whole economy, and B is the one suffering it, which is why it is (super) insured. Now A owes $100 to B and, even if there is a very very long chain of derivatives going from A to B from B to C ... all the way to Z'''', as long as $100 are collected (as all those derivative contracts are settled) and they end up in the hands of A, A can pay B its dues. To achieve this result three things are needed: (1) a total of at least $100 in either bills or "credit" must be available in the system; (2) none of the individuals involved cheats or finds itself able to grab the bills it must pass on to the next person in the chain; and, last but not least, (3) $100 of actual real nice and edible output(GNP) is available for delivery to B in exchange for those $100. These three conditions are all crucial, but (3) is especially subtle: if there is no real nice and edible output available worth $100, but just $100 dollar bills made of unedible paper, prices will go up proportionally (maybe with the long and variable lags of Milton) and B will not really receive $100. It will receive only worthless inflated pieces of green paper.

'<h' . (('4') + 1) . '>'Curious things may happen'</h' . (('4') + 1) . '>'

Interestingly enough, failure of (3) is, in our actual historical circumstances, strictly linked to the occurrence of (iii) (not (ii), (iii)) and it seems to be the way in which we (well, not me: the US Fed, the US Treasury and all those other powerful guys) have decided to pretend that (ii) happened when, in fact, it was (iii) that really tookplace. Before getting to this point, though, we will need to go through other two steps. First we will need to have an idea of what the value of outstanding derivatives is today in the US, so we can talk real numbers. Second, we will have to figure out why neither (i) nor (ii) happened, but instead (iii) did and none (none?) of us figured it out until the other day. The first part I do now, the second, which is longer, will have to wait for tomorrow or the day after. The third part will then discuss why the "solution" we are apparently heading toward may be anything but a solution. Better said, it is a solution for the few lucky As, but not for the many and unlucky Bs. Derivatives, remember, are a distributive tool and it is (gigantic) income and wealth redistribution we are talking about these days.

'<h' . (('4') + 1) . '>'Some numbers, and a puzzle'</h' . (('4') + 1) . '>'

Anyhow, for tonight let me just end with the estimates I promised. How big can the notional value of outstanding derivatives become? Very large indeed. Recall that, using my friend Ellen's careful estimates, I un-carefully valued the total amount of outstanding US real assets at around $300 trillion. In the Wikipedia page on derivatives linked above we read:

 

Over-the-counter (OTC) derivatives are contracts that are traded (and privately negotiated) directly between two parties, without going through an exchange or other intermediary. Products such as swaps, forward rate agreements, and exotic options are almost always traded in this way. The OTC derivative market is the largest market for derivatives, and is unregulated. According to the Bank for International Settlements, the total outstanding notional amount is $596 trillion (as of December 2007)^[1]. Of this total notional amount, 66% are interest rate contracts, 10% are credit default swaps (CDS), 9% are foreign exchange contracts, 2% are commodity contracts, 1% are equity contracts, and 12% are other. OTC derivatives are largely subject to counterparty risk, as the validity of a contract depends on the counterparty's solvency and ability to honor its obligations.

 

That is: just the OTC derivatives (which, as you will see, are our main concern in this saga) have a notional value that is twice my bold estimate of total US equities of any kind, which included those that do not exist because, for example and for good reasons, slavery is forbidden. If we stick to what I claim Ellen's estimate of the outstanding capital stock of the US is, the ratio is 10 (ten). To these OTC derivatives one should add the exchange traded ones, that the IBS estimated to be at around $400 trillion in 2006. Now, these numbers are for the whole world, but even if we assume that those "related to the US economy" are only 20% of the total (and they are much more) we are talking $200 trillion in notional value of derivatives, versus an actual capital stock which is about 1/4 of that and a GNP that is ...$5 trillion short to make it 1/10 of it.

Hence the puzzle: if only 1/10 of them derivatives "had to be paid" (what this means and who should pay whom, we will see in the second part of this trilogy) where the hell do we find the money? Even by loading the entire GNP of the US into the backyards of the lucky winners of the "tango game", call them Bs, we would not only be going hungry ...we would be short of $5 trillion. Peanuts, no?

Well, don't the Fed andthe Treasury own two printing presses, one for $$ and one for debt? Yes indeed, they do. Stay tuned.