Monday, November 30, 2009

Two quick notes on epistemology/philosophy of the mind

1. Steven Landsburg has come out with a book that seems rather promising. There is a blog too.

There is this one post in which he takes Richard Dawkins to task about the latter's insistence that evolution by itself destroys the idea of a God. I don't care much about the post itself, but one paragraph caught my eye.
That, however, is just wrong. It is not true that all complex things emerge by gradual degrees from simpler beginnings. In fact, the most complex thing I’m aware of is the system of natural numbers (0,1,2,3, and all the rest of them) together with the laws of arithmetic. That system did not emerge, by gradual degrees, from simpler beginnings.

If you doubt the complexity of the natural numbers, take note that you can use just a small part of them to encode the entire human genome. That makes the natural numbers more complex than human life. Unless, of course, human beings contain an uncodable essence, like an immortal soul—but I’m guessing that’s not the road Dawkins wants to take. (emphasis mine)

Has there ever, ever been a case where this phrase has been more applicable? A small part of the set of natural numbers can be used to represent the genome, not to encode it in the sense that gene encoding is commonly understood. Protein molecules can not be created out of natural numbers.

Landsburg's definition of complexity of genome in terms of encoding the genome is essentially an information-theoretic one. The main idea here is that of Kolmogorov complexity. It is the right approach, of course, but it is essential to remember just what the base information is. The genome itself is not a sequence of non-random and finite set of ACTGs, only its representation is.

Of course, there is a more fundamental though more subtle issue. Kolmogorov complexity proceeds by assuming arithmetic, which assumes the set of natural numbers. To talk of the complexity of the set of natural numbers itself is completely meaningless. It is like saying that the set of alphabets of the English language is more complex than any arbitrarily complex philosophical idea (for e.g. - the map is not the territory) because the idea can be expressed by a small, finite and non-random subset of the alphabet.

Does this mean that I'm opposing Godelian platonism by comparing a mathematical set to a human construct like an alphabet? Not really. I hold an information-theoretic view of the universe. Math is the closest we get to a system that captures the fundamental notion of information.

2. On EconLog, Bryan Caplan responds to Robin Hanson in the latest Caplan-Hanson debate (on philosophy of the mind, this time). He links to an old paper by him on why he is a dualist (mind and matter/body are different - mental phenomena can not be reduced to physical ones) and not a monist/reductionist (all things are made of matter, there is no mind beyond the body, mental phenomena are completely reducible to physical ones). He refers to John Searle approvingly before critiquing him. That's for later though.

What is striking is - Caplan claims to be neither a substance dualist (that the mind is a separate substance than the body) nor a property dualist (the mind is an emergent property of the brain that is not reducible to anything simpler). Substance dualism is rather naive and uninformed by anatomy and neuroscience. Property dualism, however, is an extremely attractive position to adopt if you're going to be a dualist. Why then does Caplan reject it? He explains -

"Neither is my view property dualism; for the essence of a property is that it could not even be conceived as existing apart from something else. For example, "whiteness" could not even be imagined to exist all by itself; the reason is that it is a property, not an independent thing. But we can conceive of the mind all by itself; hence it is not a property. "
It is an old paper, so maybe Caplan has not reviewed it in quite some time, but it is rather unbelievable that he ever used/convinced himself of the trickery of 'whiteness' (as opposed to white) and the ambiguity of 'conceive' to dismiss property dualism. 
I can absolutely conceive of 'white' by itself. You can argue that this is false, and that my conception of white will be a white wall, a white board, white cloth, brilliant sunlight or something physical that possesses the property of 'whiteness'. You will be right. But then, you can't genuinely believe that you can conceive of a mind independently. You can only conceive of mental phenomena - desire, choice, anger, pain. And even then, you can only conceive of how you/other sentient beings perceive these phenomena or how they are transient properties of sentient beings. And there's no reason to assume that something physical (the brain) cannot possess the properties implied by mental phenomena. That, in fact, is the very premise of substance dualism. 
Which is to say, Caplan rules property dualism out either by naivette or by assumption,  in very simple and direct language that makes the mistake rather obvious. 

Wednesday, November 04, 2009

On Rand and Ethics

At Marginal Revolution, Alex Tabarrok has a post on Ayn Rand's relevance and her philosophy, prompted by the recent flurry of activity around the release of two new biographies of her. He links to almost 5-yr old pieces on her by himself, Tyler Cowen and Bryan Caplan.

In his 2005 post, Tabarrok argues that Rand is a lucid, prominent and the first modern proponent of virtue ethics. Now I had no clue what virtue ethics was, so I decided to check it up. The wikipedia entry for virtue ethics says this
Virtue theory is an approach to ethics which emphasizes the character of the moral agent, rather than rules or consequences, as the key element of ethical thinking. This contrasts with consequentialism, which holds that the consequences of a particular act form the basis for any valid moral judgment about that action, and deontology, which derives rightness or wrongness from the character of the act itself rather than the outcomes
Here is the wikipedia article, which I believe is rather muddled. A clearer and crisper explanation of the differences between the three main strands of normative ethics(systems of distinguishing the right from the wrong - deontology, consequentialism, virtue ethics) can be found in this sentence from the wiki article on deontological ethics.
Deontological ethics or deontology (from Greek δέον, deon, "obligation, duty"; and -λογία, -logia) is an approach to ethics that determines goodness or rightness from examining acts, rather than the consequences of the act as in consequentialism, or the intentions of the person doing the act as in virtue ethics.
This is a very good summary of the critical difference and it enables further analysis. A lot of people will consider a 'It was never my intention to hurt you, I just ended up doing something stupid' explanation as legitimate if it comes from a romantic partner or a child. It is difficult for a legal system to adopt such a view in a case of, say, drunken driving. (Most legal systems actually hold such a view partially - we will come to that a little later.) To the extent that you can only infer a person's motives and moral character from their actions, it seems that either virtue ethics is completely subsumed under deontology or the distinction is too subtle for the gross sieve of my comprehension. In any case, though intention and action can be separated, repeated action is the surest indicator of intention that there is - you may not forgive the same mistake thrice even in a partner or a child.

I have never found deontology particularly illuminating beyond the insight that every system of normative ethics will require certain axioms. Deontology's way to distinguishing the right from the wrong is to check if the act conforms to a certain set of pre-decided rules, the moral axioms. It is easy to see that if you allow these rules to be long enough and complex enough and democratic enough, deontology collapses into consequentialism. Kant, for example, is led to conclude that lying is always bad only because he insists on working with the simple and singular predicate of 'lying'. If you allow for more complex and more numerous predicates, it is easy to accommodate lying in those situations where it creates more favourable social outcomes without explicitly mentioning that you are trying to improve the net social outcome. If you do not allow for such predicates, you are asking for the reality to conform to your thoughts, which is not a useful way to approach a theory of practical reason. Ironically, due to simpler and fewer predicates, a functioning large-scale deontological system will necessitate a much larger, more confused and less universal set of moral axioms.

Drunken driving and injury caused under the influence of alcohol is an interesting illustration. The chain of reasoning that outlaws drunken driving appears to me to be this:

1) It is wrong to cause death or injury to other people.
2) Drunkenness leads to significantly reduced motor control which is very likely to cause uncontrolled driving.
3) Uncontrolled driving is likely to cause death or injury to other people on the road for no fault of theirs.
4) Hence, drunken driving should be illegal.

A 'pure' deontological framework will find 'injury caused' a difficult concept to punish as it is a consequence and not an act itself. Thus in a deontological framework, all instances of drunken driving will appear as equal evils to you. You will not be able to engage in the kind of probabilistic reasoning I mention above, as it is fundamentally consequentalist, with the one simple moral axiom of 'don't cause death or injury to other people'.

In fact, since the intention/moral character and the consequence of an act are mutually exclusive and completely characterize the act, consequentialism and virtue ethics are better compared with each other than either with deontology. Most functioning legal systems require the establishment of both actus reus (that the act happened) and mens rea (the intention behind the act) to establish a crime. A distinction is made between drunken driving that caused injury and drunken driving that didn't, and between murder (intended, planned homicide) and the common 'culpable homicide not amounting to murder' used if a car runs over and kills someone when the driver was inebriated. Thus, we see a combination of consequentialism and virtue ethics pervading most modern legal systems.

As Tabarrok says, Rand's ideal of the virtuous man is rational, independent and productive. The insertion of rationality is brilliant stroke to achieve consistency within virtue ethics without sacrificing common wisdom, for it removes the dichotomy between evil acts and evil intents. Competence as morality brings morality closer to a folk reality. If you cause me hurt by being stupid rather than malicious, you are still in the wrong. This makes it possible to scale Rand's system up to the social level - otherwise it would remain stunted to a set of personal ethics (e.g. "yes, Salman Khan ran over and caused the death of n pavement dwellers, but he didn't want to do that any more than you or I did, so I can still think he is not a bad person and like him").

Till this point, Rand seems not only correct but also very original. However, while the moral axioms of rationality, independence and productivity are in consonance when evaluating the individual, they are less helpful in guiding a social framework. First, notice that of the three, only independence truly scales up. As Caplan points out, Rand thought that a large number of people were fundamentally irrational and led themselves to near-slavery (by supporting socialism when a 'superior, more rational, more moral' brain could see that it was doomed). She sure wouldn't want these people to be punished for being thus incompetent and immoral (beyond the market forces). It's virtuous to be rational and productive, ok to not be so competent and evil to be impinging upon somebody else's attempt to be so. And one doesn't have the moral obligation to try and make someone else rational and productive. How different is this from the concept of negative liberty, an idea that has had currency at least since the time of Thomas Hobbes?

Therefore, I believe Tabarrok missed the point when he asserted that Rand's moral defence of capitalism on the basis of selfishness is spot on, for a much more satisfying defence is the simple and older idea of negative liberty. Rand was also neither unique nor the first in requiring that there be a moral defence of capitalism - this had already been done explicitly by Frank Knight, and implicitly by others who identified as classical liberals. Tabarrok was also grossly unjust to every single political philosopher of the modern times when he says that "the modern literature lags behind Rand in connecting ethics and politics". All political philosophers are first moral philosophers. It is impossible to try to create a coherent governance framework without first privileging certain social conditions and outcomes as moral 'goods'. Political philosophers don't make their ethics explicit, because they are usually building upon or deconstructing the politics (and hence ethics) of older political philosophers. Isn't the Hobbesian Leviathan predicated upon the assumption that human life should be neither nasty nor brutish nor short?

It is because of all this that I tend to judge Rand's moral philosophy rather unfavourably. Her insistence on selfishness and egoism as opposed to mere negative liberty opens weak holes in what is otherwise a sound philosophy. Her insistence on evaluating the individual and then creating a society that best aids the most perfect individual lays her philosophy open to the charge of ignoring the social reality and of not being universal enough to merit a discussion as a political system. If you respect the political equivalence of all individuals (everyone must have the same rights as I do) it is disingenious and contradictory to arrive at that set of rights based upon your preference and your own ideals without even considering the preferences and ideals of those whom you want to grant those rights. It is difficult to see what Rand had to offer that may be relevant to any scaled-up political discourse that Thomas Hobbes, John Locke or John Stuart Mill did not.

And it is the intuition that democratic rights be democratic in nature that makes consequential and utilitarian ideas so attractive. You don't need to have a personal ideal of the perfect man as independent and productive and rational to arrive at a political system. You only need the intuition that the average man desires to be free, to be materially prosperous and to be able to make a better decision to arrive at the notion of a social utility function. It is only a small step from there to the idea that a political economy and a legal system that makes decisions that maximize the social utility is the ideal system. And some more small steps with mathematical reasoning will take you to Pareto optimality and Kaldor-Hicks efficiency.

Of course, the version of utility and efficiency that we have right now is very broad - it includes the human preference for liberty and even the human aversion for inequity. Therefore, modern political and economic philosophy explicitly recognises the trade-offs between efficiency, liberty and equality. As an illiberal society is also usually an inefficient society and as the preference for equality can be accommodated to a broad extent in macro aggregates (Gini coefficient, Human Development Indices) it is eminently possible to zero down upon a political system that is quite close to maximizing social utility - the modern liberal democracy, capitalism with a few rules. The debate then becomes about how many of those rules should be there and what those rules should be. Rand doesn't have much to add there.

Of course, in the last two paragraphs I have casually slipped in the idea of positive liberty through efficiency and equality arguments. You can disagree. However, if you must try to create a grand theory to back your belief in laissez-faire, you are much better off looking towards negative liberty and the classical liberals than selfishness and Ayn Rand.

p.s. On epistemology, though, Rand is surprisingly good. From Caplan's post
"We know that we know nothing,' they chatter, blanking out the fact that they are claiming knowledge–"There are no absolutes," they chatter, blanking out the fact that they are uttering an absolute–'You cannot prove that you exist or that you're conscious,' they chatter, blanking out the fact that proof presupposes existence, consciousness and a complex chain of knowledge: the existence of something to know, of a consciousness able to know it, and of a knowledge that has learned to distinguish between such concepts as the proved and the unproved.
If you forgive the subtle fallacy of recursion without base class in the first two sentences (it is perfectly logical, for example, to say that you believe in tolerance and hence are intolerant of intolerance), this is a brilliant paragraph. The "proof presupposes ...." part is sheer genius in its clarity and delivers a short yet fatal blow to a lot of Cartesian and Humean skepticism.

p.p.s My bias towards efficiency and utility, while sympathizing with negative liberty and the human aversion for inequity at the same time may seem contradictory. It is. I have the broad framework clear, and this framework is essentially that of John Stuart Mill. Within that, I prefer to think about the specific tradeoffs on a case by case basis and using inductive logic. If you insist that a social system be logically consistent and yet be able to handle all 'truths' that intuition and reality will throw at it, you have missed Godel's fundamental insight.

Tuesday, October 06, 2009

Liquidity Trap, Banks & Fiscal Policy

Monetary policy is usually considered potent enough to cure all evils. If the economy is not growing fast enough - the central bank buys treasuries for cash from banks, thus releasing more money into the system for circulation. The economy is overheating, it does the reverse. Equivalently, it can also lower or hike interest rates by fiat. A decision by the central bank to buy treasuries for cash or to decrease interest rates has the same effect - it increases the price of the treasuries and makes it desirable for banks to sell them and invest the proceeds elsewhere.

All this assumes a normal economy where a bank has no incentive to hold reserves in excess of what it is legally required to do and where treasuries and cash are not good substitutes for each other. In a crisis, the economy is not in a normal situation. Interest rates could be touching zero (as they are now) and treasuries and cash become almost substitutable. If the central bank now starts buying treasuries for cash from banks, the banks will happily take the cash and sit on it. They will keep reserves much in excess of what they are required to, as they don't have confidence in private assets, treasuries are equivalent to cash and the possibility and fear of a bank run (or a loss of confidence in the banks' assets) makes it prudent to keep more reserves than is mandatory.

The part where banks consider the excess cash as a substitute for treasuries and hold on to it (just like they were holding on to treasuries) is the liquidity trap. The part where banks hold excess reserves due to fear, let's call that a 'solvency trap'.

To get past this situation, Keynesians recommend fiscal stimulus. The government issues new treasuries, this increases the interest rate on treasuries, people buy treasuries (through banks) and the government has cash to burn. Of course, some of the excess cash will also be used up to buy existing treasuries from the central bank itself, i.e fiscal stimulus will also have the effect that a mild monetary contraction in the form of an interest rate hike does. This can be avoided by concurrent monetary expansion by the central bank.

Irrespective of any such concurrent monetary policy measure, fiscal expansion ensures one critical thing at least - the government borrows from lenders because the private sector is not able to. The average willingness to spend cash has increased. The government can then spend the money in various ways. It can choose to invest in infrastructure and aim at the positive network effects to boost the economy. It can reduce the debt burden of households and businesses by re-capitalizing them. Or, it can decide that it has no competence in choosing and privileging specific households and businesses and instead just aim at getting the stimulus to work through the banking system by giving cash back to the banks.

But lack of cash with banks was never the problem - the willingness to get it moving was! And the government is not going to just make a donation to the banks. So what happens in a fiscal stimulus that operates through banks anyway?

Well this is what I think happens. Government issues new debt, lenders of all hues move cash from banks to treasuries. The balance sheet of the banking system shrinks. The banks also buy up some more treasuries from the central bank. Now the government is going to give this cash back to the banks in exchange for something. What is this something?

It won't be treasuries, that will just be a monetary expansion and fiscal contraction. It won't be true short term debt - that's just cash back to the bank and all that has happened is that the bank owes money to the government instead of investors and the government in turn owes money to the investors. It can, however, be long-term debt, or short term debt with the surety that it will be rolled over (which is just cheap long-term debt) or equity. If it is long term debt, the asset liability mismatch of the bank is corrected to an extent. If it is equity, the bank is unlevered a little without a shrinking of the balance sheet. In either case, fiscal stimulus that operates through the banking system fundamentally improves the capital structure and hence the solvency of the banking system. This is bound to reduce the tendency to hold excess reserves and get the money circulating faster.

Conventional monetary policy then just involves a shift between two safe assets of the banking system - it cannot be expected to work when there is substitutability between these assets. Unconventional monetary policy, in which the central bank buys private assets from banks in exchange for cash, also exchanges one asset for another but these are not substitutable. It can, in principle, work.

Bank-driven fiscal policy effectively involves tinkering with the liability structure of the banking system. It can be a way out of the liquidity trap as well as the solvency trap. In fact, it can help the economy escape out of the liquidity trap precisely by getting out of the solvency trap.

Tuesday, September 22, 2009

Promised Econ Link Fest

1) The Blog Wars : De Long & Krugman vs. the New Classicals

De Long takes apart the New Classical case: here, here and here

2) Nick Rowe explains why money is not like other goods and assets and hence, not neutral in a non-barter economy, here and here.

Keen is a Minsky style Post-Keynesian. He has written a book called Debunking Economics, which is highly critical of neoclassical economics.

He laments the usual equilibrium analysis of the economy in the mainstream and tries to model the economy as a non-linear dynamical system using differential equations. He argues that money is endogenously created (money is there iff. there is debt). He also sees the modern capitalist economy as inherently unstable (which ties up with the dynamical model) because of the financial fragility hypothesised by Minsky.

4) Barry Eichengreen and Kevin O Rourke on the comparison between this depression and the original great depression.

Yellen is the governor of the Federal reserve of San Francisco. She argues that there is a pressing need to incorporate finance into mainstream macro, that there indeed are asset price bubbles and it may be desirable for a proactive central bank to deflate them. She also says that not all bubbles are equal, and that credit market bubbles and deleveraging need more proactive intervention than equity market booms and busts. She is however skeptical about how a central bank can go about doing it, and suggests countercyclical capital requirements (which go up in times of liquidity and go down in times of distress) as a robust policy measure.

6) 2006 paper by Alan Blinder on the main monetary policy challenges faced by central banks everywhere.

Blinder's only blind spot in the paper was his strong assertion on the issue of asset price bubbles. He asserts that central banks should not be concerning themselves with such bubbles, and gives the example of the tech bubble and bust to demonstrate how asset price fluctuations can be effectively handled post facto. In the light of recent events, it is a particularly bad example. Oh well, hindsight is perfect.

The insights on exchange rate interventions (intervene in extreme volatility), transparency of output and unemployment objectives (its necessary) and bank supervision (it may be required), and the deliberations on interest rate are quite brilliant. Do read the whole thing though - it's clear, smoothly written and really gives the lowdown on the practice of central banking.

7) Minsky & Minsky-like

What can I say that hasn't been said before? From wiki,

Dr. Minsky proposed theories linking financial market fragility, in the normal life cycle of an economy, with speculative investment bubbles endogenous to financial markets. Minsky claimed that in prosperous times, when corporate cash flow rises beyond what is needed to pay off debt, a speculative euphoria develops, and soon thereafter debts exceed what borrowers can pay off from their incoming revenues, which in turn produces a financial crisis. As a result of such speculative borrowing bubbles, banks and lenders tighten credit availability, even to companies that can afford loans, and the economy subsequently contracts.

This slow movement of the financial system from stability to crisis is something for which Minsky is best known, and the phrase "Minsky moment" refers to this aspect of Minsky's academic work.

Paul McCulley's Minsky inspired take on the 'shadow banking system', a term that he coined. Incidentally, Janet Yellen also refers to McCulley approvingly in her address. Oh well, PIMCO is just a great bond fund.

8) The history, development, state and relevance of macro (the non-vitriolic not-just-post-crisis series),

David Laidler, among the original monetarists, on Lucas, Keynes and macro.

Robert Gordon, on why 1978 era macro might be the best guide to the crisis. Some insightful ideas about the differences in goods that have 'auction markets' (like oil) and others that don't and what that means for macroeconomic theory.

Greg Mankiw, 2006 paper on theory vs. policy in macro. Excellent stuff!

Arnold Kling, in a smart little piece about how shortage of data affects macroeconomic debates of all hues, as a small review of Mankiw's paper.

Michael Woodford, papers on revolution and evolution in 20th century macro (1999), and a somewhat ill-fated celebration of a then new neoclassical 'consensus' in macro. (2007)

Robert Lucas, asserting that smoothening short run demand flucatuations ought to be subservient to looking at long term supply driven growth. This address forms the basis of part of the analysis in the Laidler paper. Recent events don't bear him out too well either.

Have fun!

Saturday, September 19, 2009

Money, Macro & the Crisis

Recently, I've found myself reading a lot on the crisis, monetary policy and macroeconomics. This includes the usual suspects, some newly discovered stuff and a book. But most of all, I've been reading research papers, including some that were published a few decades ago. Some sort of clarity and a framework have begun to emerge and I'm now finally able to get the drift on some of the more technical discussions on the relevant issues. There are two prominent themes in these readings that emerge often and that I find myself agreeing with most strongly.

The first is that there is a pressing need to incorporate finance and into macroeconomics. Earlier, this incorporation had stopped at understanding the role of money in the economy with a cursory glance at the banking system. The basic channel that unifies money, the banking system and the real economy is credit and credit has never really been incorporated into mainstream macro. If there is only one thing that we learn from Minsky, it should be this: incorporate credit as the concept and the financial sector as the context to inform macroeconomic policy.

The second theme is that dynamic monetary disequilibrium is a consistent feature of the modern credit economy. If you take a look at the criticisms of the New Classical school (esp. John Cochrane) coming from De Long and Krugman and many others in the recent online economic wars, there is one constant refrain: the New Classical school has failed to grasp the role that money can play in the modern economy. The New Classical school has an implicit belief in the Say's Law, and thus believes that there can be no general supply glut, that excess supply in one sector is offset by excess demand in another, that savings always equal investment and hence recessions are temporary readjustments in overall social preference from some goods and sectors to other. It thus believes in the 'neutrality of money', i.e the idea that money only functions as a medium of exchange, that investment and consumption decisions are made in real terms and hence the only source of demand for money is the 'transactions demand'. As many have pointed out, this is a model that could perhaps accurately describe a barter economy, or to a monetary economy where the velocity of money is a fixed technological constant subject only to exogenous shocks and thus behaves as if it were a barter economy.

While the barter system has not existed for a long time now, the relatively simpler economy of the past, say until the 19th century, could perhaps be approximated by such a model. In an explicitly monetary economy, however, money also seems to perform the function of a store of value and thus there could be disequilibrium, especially a monetary disequilibrium. A fully fledged 'finance economy' has since become the reality in the 20th century, and such a finance economy seems to have the ability to magnify and reinforce the monetary disequilibriums.

There are, of course, a number of people and streams of thought that have tried to incorporate credit into their understanding of macroeconomics and business cycles. Apart from Minsky, the Austrian Business Cycle theory is one such. Broadly, it says that money is easy in good times due to actions of market participants and artificially low interest rates set by the central bank. This excess money causes businesses to over-invest, thus temporarily causing a bubble. Ultimately this bubble bursts, resulting in a temporary under-investment that brings the economy back to its 'normal' state. Central banks should not try to increase liquidity during such a crisis, for that will just create another such bubble. Murray Rothbard's version of the business cycle also has people over-inesting irrationally only in the presence of the central bank!

Because of the resemblance of the boom-bust hypothesis to the technology bubble to mortgage bubble shift of the past decade, and because of its incorporation of credit, this theory has gained some credence in the past year. It has however, been criticized on various grounds at various times by several economists, including both Keynes and Friedman. The central issue with the theory is that it also seems to believe in the neutrality of money. Moreover, the theory seems to treat irrational over-investment as a moral evil and recessions are seen as having the salutory effect of bringing people back to the preferred state of rationality. This can seem as a Great Depression-apologist stance and Krugman has indeed attributed part of the blame for the broad-based central bank failures during the depression to the Austrians.

Apart from the Austrians and Minsky, there was at least one other economist whose ideas seem strikingly close to the events unfolding of the crisis, but who doesn't seem to have gotten much attention. Henry Simons is considered one of the founders of the Chicago school. Simons recognised and distrusted the high volatility of the liquidity demand of money and believed that the credit cycle and short term debt issued by banks and corporations exposed any financial system built on it to severe fluctuations through asset liability mismatches, which could then spread the fluctuation to the real sector. He was also concerned a lot about maintaining the price level. In one theory, we see the amalgamation of several strands of thought that have come to dominate academic thought in the crisis.

Though Simons's understanding of the inherent fragility and positive feedback of the financial system is similar to Minsky's and Bernanke's, his policy prescriptions differ. He recommends full reserve banking, the gold standard, the complete absence of short term debt, and strict monetary management by the government. If these frameworks are in place, it seems to me that the rest of the economy starts resembling an 'as if' barter system and his favoured policy stance of laissez faire becomes workable.

A full reserve banking solution has also been provided by others wary of the financial system or the government, including Rothbard and intermittently, Friedman. The causes of course, differ. The generation of money through fractional reserve banking has been compared to counterfeiting by more than a few respected economists, including Maurice Allais. The debate is not only interesting (though perhaps lop-sided), it is critical to understanding the tenuous relationship between money and credit in the financial economy.

Credit is endogenous, it comes about from the interaction of the financial and the real sectors. Indeed, it is also the main channel of transmission between the two. Proponents of fractional reserve banking argue that it is just the way to implement the 'monetization' of credit. Proponents of full reserve banking are concerned either about the supposedly illusory money (Rothbard, Allais) or about financial fragility (Simons).

Of course, even within the 'monetization of credit' philosophy, one can ask where does the money itself come from. While the monetarists and many Keynesians believe that money, though inevitable, is still exogenous and comes from the government and the central bank as a way to monetize the credit, the circuit theorists argue that money is an endogenous property of a credit economy itself. Indeed, in their view, credit IS money. Of course, there are the agnostics in between these two views.

We can thus summarize the broad ways of looking at the theory and concept of money, arranged in a roughly free market to interventionist order.

1) Money is neutral and always in equilibrium.
2) Money is neutral but there are temporary, required credit disequilibria.
3) Money is not neutral and sometimes in disequilibrium, but exogneous.
4) Money is not neutral and often in disequilibrium, and may be exogenous or endogenous.
5) Money is not neutral, often in disequilibrium and endogenous.

My own views are closest to 4.

(This post has left out several contextually required links. The next one is going to be a mega link fest)

Sunday, September 13, 2009

RBI and central banking

If you read Ajay Shah's blog, you will notice that he is an ardent supporter of inflation targeting, rule based central banking and by and large critical of RBI's policy actions. Specifically, he believes that in an emerging country likely to have weak institutions, like India, the central bank has no business using discretion, that currency depreciation should not be fought using fx reserves, that modern academic macroeconomic insights (of the Monetarist-New Keynesian variety) like inflation targeting are widely used and widely effective and that RBI's heavy handed approach to regulation has drastically stunted financial markets and innovation and thus stunted entrepreneurship and growth. Ila Patnaik, with whom he has collaborated on several papers and other research initiatives, has more or less similar positions. I refer to the two of them because of two recent pieces - this op-ed by Dr Shah in FE and this one by Dr. Patnaik in the Indian Express.

Before we turn to the op-eds in question, it is interesting to explore each of the broader positions in some detail. Rule based central banking is thought to stabilize market expectations and prevent the central bank from committing policy mistakes which it might otherwise do on account of insufficient data, incompetence or vested interests. However, it also takes away some important tools that central banks have to fight pro-cylicality. Dr. Shah is however ok with de facto if not de jure central bank positions, but seems to insist that India cannot afford de facto positions because our institutions are weak. This characterization of Indian institutions, while broadly correct if we were only talking in generalities, fails as a logical argument when one specific institution is being discussed, whether the RBI or any other. Rule based central banking could be a great idea if there was sustained evidence of RBI policy failure, otherwise there is no specific need for us to be pessimistic about RBI simply because on the average, Indian institutions are weak. And how has the RBI done? We'll turn to that in some time.

The specific measures of inflation targeting and using fx reserves to fight currency depreciation are trickier issues. Basically, as this article by Raghuram Rajan points out, the two objectives cannot be managed well together with a single instrument - interest rate - at your disposal and currency defences are widely thought to be ineffective at best (as in the case of England, 1992) and ruinous (Indonesia & Malayasia, 1997) at worst. Combine that with the hypothesized effectiveness of the Taylor Rule in combating stagflation, and we seem to have a no-brainer: open capital account completely, let the currency fluctuate and manage the price level through inflation targeting. Broadly, this makes complete sense. However, there are important nuances that are probably being missed out in such a simplified narrative.

For one, emerging market currencies tend to be strongly correlated with all other asset markets in the country. Most capital inflow into emerging markets is on the basis of perceptions of macroeconomic and political risks and stability. In India, the situation is further complicated by the fact that foreign institutional investors are not allowed to buy and sell rupees in the spot (cash) market, unless it is for the explicit purpose of buying equities or some other assets. Now anyone who has any knowldge of trading desks will tell you that such correlations are often speculated on, or other speculations are often based upon assumptions of such correlations. In a crisis, when there is sudden asset deleveraging, such speculations can form a strong postive feedback looop, further aggravating the deleveraging. It is futile to try to establish the cause and effect chain in such feedback cycles - to counter them, one or both the legs of the cycle may be attacked, and this could be the fx market or the correlated asset market (equity, for example). It is interesting to note that the darling of the free market, Hong Kong, had managed to avert the 1997 crisis to a great extent through such operations. The nature of the operations may be different for an economy like India, for the degree and even the direction of the correlations may run different, but point is - it is possible and perhaps desirable to defend a currency provided one does so judiciously. And without discretionary monetary policy, such actions are impossible.

What of inflation targeting? Though the evidence has been mixed, IT gives a robust way to achieve price stability, which is key in an economy with fiat currency, fractional resevrve banking and credit. The question is not whether IT is a worthy goal - it surely is. The question is - should IT be the only focus of the central bank, or are there other objectives just as worthy? Specifically, should the central bank pro-actively intervene in asset price bubbles? This paper by Janet Yellen (governor of Federal Reserve of SSan Francisco and one of the economists touted as Bernanke's successor) discusses the issue in great detail. Yellen notes that not all asset price fluctuations are created equal and broadly says that while equity market booms and busts may be left alone, credit markets deserve a deeper look as credit is the main mechanism of transmission between the financial and the real economy. In either case, praising and echoing Minsky, she emphasizes the need to bring the financial sector and financial stability as explicit constructs in macroeconomics and monetary policy.

Next, what about capital account liberalization? Let's treat the restrictions on corporate debt investments later. First, let's recall what exactly is forbidden in the Indian currency market - FIIs are not allowed to speculate on the spot value of the rupee. They can buy and sell NDFs (non-deliverable forwards) though, and these are short-dated enough to be considered close proxies for the spot market by most trading desks. What this ensures, however, is that the very short term rupee liquidity in the market is insulated from manpulation, allowing RBI greater room to act as a manipulator itself. Raghuram Rajan, whose committee has recommended full capital account convertibility, himself finds in this paper that
Cross-country regressions suggest little connection from foreign capital inflows to more rapid economic growth for developing countries and emerging markets. This suggests that the lack of domestic savings is not the primary constraint on growth in these economies, as implicitly assumed in the benchmark neoclassical framework.
and goes on to suggest a 'pragmatic' way for the process of capital account liberalization. In another paper, he and his co-authors conclude that
even successful developing countries have limited absorptive capacity for foreign resources, either because their financial markets are underdeveloped, or because their economies are prone to overvaluation caused by rapid capital inflows.
seemingly confirming some of the common sensical arguments around the 'hot money' of capital inflows.

Now finally, we turn to the recent critiques that Dr. Shah and Dr. Patnaik have dished out to the RBI. Dr. Patnaik asserts that India is not the only country that has remained relatively unscathed in the crisis and hence there is no evidence that RBI has done a spectacular job and that the world has nothing to learn from RBI. She criticizes the self-praise that some recent RBI establishment people have heaped upon themselves (perhaps Dr. Y V Reddy) and points to the micro and macro-prudential frameworks of east Asia as being worthy of emulation. She concludes memorably
A villager with no roads may foolishly boast of having no accidents, but he cannot teach people how to regulate traffic on busy intersections. It is important for policy makers to remember that India has no lessons to offer to regulators operating in the sophisticated world of finance, and proposals suggesting that they should learn our style of regulation only makes us look foolish.
Let's take a look at what RBI and Dr. Reddy have praised themselves for. This NY Times article provides a clue.
One of the first moves he made was to ban the use of bank loans for the purchase of raw land, which was skyrocketing. Only when the developer was about to commence building could the bank get involved — and then only to make construction loans.
Then, as securitizations and derivatives gained increasing prominence in the world’s financial system, the Reserve Bank of India sharply curtailed their use in the country. When Mr. Reddy saw American banks setting up off-balance-sheet vehicles to hide debt, he essentially banned them in India. As a result, banks in India wound up holding onto the loans they made to customers. On the one hand, this meant they made fewer loans than their American counterparts because they couldn’t sell off the loans to Wall Street in securitizations. On the other hand, it meant they still had the incentive — as American banks did not — to see those loans paid back.

Seeing inflation on the horizon, Mr. Reddy pushed interest rates up to more than 20 percent, which of course dampened the housing frenzy. He increased risk weightings on commercial buildings and shopping mall construction, doubling the amount of capital banks were required to hold in reserve in case things went awry. He made banks put aside extra capital for every loan they made. In effect, Mr. Reddy was creating liquidity even before there was a global liquidity crisis.
Now, if these aren't micro and macro-prudential measures, I don't know what are. I fail to see why Dr. Patnaik feels that these measures are something that no one can learn from. Interestingly, Yellen also notes that a way for central banks to induce financial stability is to manage liquidity requirements
Capital requirements could serve as a key tool of macro-prudential supervision. Most proposals for regulatory reform would impose higher capital requirements on systemically important institutions and also design them to vary in a procyclical manner. In other words, capital requirements would rise in economic upswings, so that institutions would build strength in good times, and they would fall in recessions. This pattern would counteract the natural tendency of leverage to amplify business cycle swings—serving as a kind of “automatic stabilizer” for the financial system."
Now compare this with D V Subba Rao's speech, the one that has come under fire from Dr. Shah. Subba Rao, outlining the measures that the RBI recommends to stave the crisis, says
24. It may be relevant to highlight some of the specific features of our system that have contributed to financial stability:
• Banks are required to hold a minimum percentage of their liabilities in risk free government securities under the statutory liquidity ratio (SLR) system. This stipulation ensures that banks are buffered by liquidity in times of stress.
• We managed the capital account actively. In the face of large capital inflows during 2006-08, we sterilised the resultant excess liquidity through calibrated hikes in the cash reserve ratio (CRR) and issue of market stabilisation scheme (MSS) securities. When the flows reversed during the last quarter of 2008, we reversed the measures too. We cut the CRR and bought back the MSS securities to inject liquidity into the banking system.
Seems like Subba Rao and teh RBI, through design or default, have integrated Minsky and Yellen's insight into the system already. He also outlines the fundamental philosophy of RBI's monetray policy and management
20. In contrast to the minimalist formula of ‘single objective, single instrument’, the conduct of monetary policy by the Reserve Bank has been guided by multiple objectives and multiple instruments. In general, our three main objectives have been price stability, growth and financial stability, with the inter se priority among the objectives shifting from time to time depending on the macroeconomic circumstances.
The RBI's philosophy then, is nothing more than what Minsky suggests in his financial instability hypothesis. It is ahead of its time in explicitly recognizing the importance of financial stability as a goal beyond price stability, though I consider Yellen's skepticism about central bank intervention in asset price bubbles healthier than Subba Rao's conviction.

By stark contrast, Dr. Shah's and Dr. Patnaik's op-eds seem like rants that assert what they want to argue. They smirk at RBI's gloating, but this is what Subba Rao says
Sure, we have been hurt by the crisis, but much less than most others. It will be a folly though to let that lull us into complacency and to believe that there is something inevitable about India’s financial stability.
Now let's be charitable and remove all the smirking and rhetoric in the op-eds and look at RBI's philosophy on the main issues - capital account liberalization, financial sector regulation, inflation targeting and currency defence. The RBI has a medium term inflation target of 5% and intervenes in the fx market only when there is "excess volatility'. RBI let the rupee fall from 40 to 52 against the dollar in the crisis, intervening only in most difficult phases. Fx reserves have not been depleted by any significant amount, and as Prof. J R Varma points out, the sterilizing of capital inflows that RBI did during the boom years meant that it was long US treasuries and short Indian equity, an immensely profitable trade in 2008. And this is what DVS says about the capital account liberalization process
We view capital account liberalisation as a process and not an event.
Indeed, this resonates with Rajan's paper. Where the RBI has really been messing up is the regulation of the corporate debt market, which irrespective of its relation to economic growth, is a trade that loses India lots of money, pointed out by Prof. Varma in the same post. This is also the ground on which Subba Rao's integration of financial stability into monetary policy objectives seems the weakest
The policy framework encourages equity flows, especially direct investment flows but debt flows are subject to restrictions which are reviewed and fine-tuned periodically.
If Dr. Shah and Dr. Patnaik were to restrict their critiques of the RBI on this account, and in a juicier more content-filled manner, one would feel bound to agree with them. However, they dilute their strong arguments with a plethora of weak ones, and rather than show some effects of the stunting of the corporate debt market (as Prof. Varma does), take potshots at the RBI's 'gloating'.

This is a classic policy debate situation. The central argument that Dr. Shah espouses is made to look weaker than it is by his undercooked takedown of his intellectual opponent.