Twitter: Joshua Gans- @jodiecongirl Actually, it says more about you that you've heard of them!
- @jodiecongirl I think it says that economists don't just assume rational agents but are attracted by them too. Talk about a value conflict.
- @jodiecongirl Why uninteresting? Think of Alex P. Keaton, Quark , Sheldon and Dogbert.
- RT @freakonomics: Freakonomics Movie is released via iTunes tomorrow: http://bit.ly/9YveA3 [I'm gathering the kids for a Friday night movie]
- Am I to understand that we are now in the path of a Hurricane? http://bit.ly/daitO9 This aint the tropics.
Readership
Nov
13
Tobin tax: Financing Australia’s current account deficit
November 13, 2009 | 10 Comments | Sam Wylie
On Wednesday I wrote that people who advocate a Tobin tax are revealing that their understanding of risk is at the level of static risk management rather than higher level of dynamic risk management. Let me use an example to explain what I mean.
Australian banks make more loans than they can fund by raising deposits or purchasing funds in Australia. So, to make up the difference, they issue a lot of foreign currency bonds into global debt markets. The total volume of extant off-shore bonds issued by Aussie banks is about $A320 billion, mostly in Euros and USD. The banks convert the funds borrowed off-shore into $A (without paying a Tobin tax) and lend it into the Australian economy. Off-shore borrowing by Australia’s the four big banks finances most of Australia’s current account deficit.
The Australian banks then have $A assets and USD and Euro liabilities. They face the obvious risk that the $A will depreciate against those currencies. The banks have to hedge that risk. The capital adequacy rules for commercial banks (the Basel II agreement) make it infeasible for banks to hold large amounts of forex risk on their balance sheets. So instead, the banks pass the forex risk of their liabilities into the risk markets, where it is sliced up and distributed to the parties who can bear it at a lower cost.
To hedge their exposure banks must lock in the forward exchange rate. There are many instruments and strategies for this purpose. The bank could use forex swaps, forex forward contracts, forex futures, forex options, interest rate swaps, T-Bond futures, etc., etc. Lets say a bank opens a five year forex swap in which the bank pays the $A 90 day bank bill rate and receives the $US 90 bank bill rate (I am not going to sweat the institutional details here too much).
If the demand and supply in this swap market was balanced then there would be as many parties wanting to swap $A for $US as those wanting to do the reverse, that is swap $US for $A. But because of Australia’s large current account deficit there is not a balance. The banks have to induce counter-parties to pay $US and receive $A. The banks do this by paying a premium (a basis) over the $A 90 bank bill rate. Historically banks have to pay an extra 15 bp over the $A bill rate to get the deal done, but lately that basis has been about 30 bp, for a variety of reasons.
The extra 30 bp induces a counter-party to come forward and accept the risk of $A depreciation from the Aussie Bank. The counter-party that pays $US then has 30bp to work with. 30 bp is a lot of love to share around. The job for the counter-party is to slice up the forex risk and pass it into the market, inducing other parties to accept the risk and still have some of the 30 bp left over. (Aussie banks can of course do this for themselves. )
The process of slicing up the forex risk and distributing it among many economic agents who can bear it at a lower cost (and receive part of the 30 bp in compensation) is the processing of trading instruments. Hedging of forex risk leads to trading in other markets.
Forex risk and interest rate risk are inseparable (interest rate parity). The risk faced by the bank of falling $A, can just as easily be characterised as a risk of relative movement of the $A and $US yield curves. So, the risk makes its way into the interest rate risk markets. Moreover, forward markets and spot markets are tightly joined by arbitrage relationships, so trading the forwards markets induces trading in spot markets — especially the spot market for forex.
The parties who absorb the risk will deal with it in a dynamic way. If the market moves against them ($A falls) then they may wish to reduce their exposure to limit losses, and that dynamic adjustment involves a lot more trading in spot and forward instruments.
The point of this example is as follows:
1. Activity in the real economy leads to a demand for credit (investment and consumption). Aussie banks meet that demand by importing capital which exposes them to forex/IR risk.
2. The banks pay to transfer the risk. Risk and debt markets slice up the risk and pass it to many other agents.
3. A large volume of trading is induced by intial hedging of the risk in spot and forward forex and IR markets. Dynamic management of risk by the multiple parties holding the risk leads to more trading and more transfer of the risk. Outsiders will observe large volumes of trading and may think it is all speculation.
4. But much of the trading volume is the financial system working the way we want it to work. Financial intermediation is meeting the demand for credit. The risk and debt markets are moving the risk out of the intermediaries and acheiving efficient dynamic allocation of risk.
5. This process, which is so crucial for Australia’s economy, only works efficiently because a massive volume of trading can be conducted at very low transaction costs.
Is increasing those transaction costs with a Tobin tax really a good idea?
Comments
10 Responses to “Tobin tax: Financing Australia’s current account deficit”
“Is increasing those transaction costs with a Tobin tax really a good idea?”
]
I dont know, what is the cost from that volume which is speculation? Would i be willing to raise the MC and reduce efficient risk transfer in order to prevent damage from overshooting markets driven by speculation…maybe…
Advocating a tobin tax doesnt necessarily imply a person doesn’t get it. Just that they see a different cost benefit trade off, no?
[disclosure: I'll admit i dont have huge depth on this stuff despite having the advantage of having been exposed to Sam's world class teaching
You again beg the question of showing that the markets are in fact “acheiving [sic] efficient dynamic allocation of risk”, and not simply speculating upon their own internal dynamics. You simply assert that “The parties who absorb the risk will deal with it in a dynamic way”.
What if they deal with it instead by (as seems to have been the case recently) buying financial products they don’t understand in the belief that they can always on-sell them in a rising market, and if the market falls then they can use political influence to have the tax-payer bail them out?
Your assertions are easily put to an empirical test, which they fail. If markets efficiently allocated risk, then they should conform to the “random walk” hypothesis in their movements and huge climbs and plunges in market prices should be vanishingly rare events. In fact, bubbles and crashes happen all the time (so called “fat tailed distribution” of events), certainly far more often than the normal distribution of movement predicted by a random walk gives grounds for. (Peters 1994, Fractal Market Analysis).
A historical test is also possible. Historically speaking, the size of the financial sector has grown massively since the general abandonment of fixed exchange rate regimes and the keynesian consensus post the 70s oil and inflation shocks. If the financial sector was so great at efficiently allocating capital and risk, one would expect this to be reflected in better economic growth since then. In fact, as Paul Krugman has been documenting in a recent series of blog posts (starting here: http://krugman.blogs.nytimes.com/2009/11/05/the-lost-generation/) precisely the opposite is true; we have never equalled the growth of 1950-1975, when the financial sector was tiny and dominated by oppressive government regulation.
More specifically, given that banks are quite capable of paying a 15-30 bp premium without suffering losses, a Tobin tax, which is usually suggested at somewhere between 1 and 20 bp, would not significantly distort this important market, but would drive out excessive speculation (see Dean Baker’s article: http://www.cepr.net/index.php/op-eds-&-columns/op-eds-&-columns/a-financial-sector-small-enough-to-drown-in-a-bathtub/). The general point is that a tobin tax penalises rapid movement of funds (a characteristic of short term speculation) far more than the kind of risk-minimisation and productive investment that you are talking about. It would thereby reduce the very volatility that the banks need to hedge against, and would be a net good for international capital investment.
Simon — If you took my Futures and Options class as well as my Banking class then you would be guru on all this stuff. I am not sure what the costs of speculation are, because I am not sure how to define specualation, or how to know which trades are speculative. If I accept a positive return to absorb risk — is that speculation?
James — thanks for the long reply. I am not contending that financial markets are acheiving ideal levels of allocative efficiency. My point is that reallocating risk, created by real activity, takes a lot of trading because of the dynamic nature of risk management. Perfectly efficient allocation would require zero transaction costs and an infinite amount of trading.
The more efficient the allocation of risk, the more trading is necessary. So, when looking at the massive volumes of trading going on in risk markets, so people want to conclude that it is all speculation, but actually those volumes are completely consistent with efficient allocation.
It is true, and well established, that trading in financial markets creates noise, and that noise causes other agents to trade. See the large literature on noise trading. Brett Trueman’s 1994 is the best starting point. However, there is little in empirical or theoretical studies of noise trading to suggest that it is destabilising in markets.
Next point, you post seems to confuse notions of efficiency. It is unfortunate that economics uses the word efficiency to mean: “informational efficiency”, “allocative efficiency” and “technological efficiency”. These three concepts are closely related, but distinct. When I say allocative efficiency, I don’t mean information efficiency.
Finally, I would urge you to come up with a good example to shows how a Tobin tax would improve aggregate welfare. Examples are great, otherwise the discussion goes around in circles.
I haven’t read Krugman’s post. But let me point out that growth in GDP is not the sole aim of the financial system. At any level of GDP, better allocation of risk across agents will cause a welfare improvement. Volumes have increased massively since 1973 because of improvements in asset pricing technology and regulation. With those technology and regulation improvements have come better allocation — regardless of what GDP was in those periods.
Cheers, Sam
The problem is that the same instruments used to lower risk are also used to increase leverage for speculation. The promoters of the Tobin Tax are trying to reduce damaging asset price swings which they believe are worsened by speculation, but at the same time this would increase the cost for responsible companies to reduce risk.
I think a better approach would be to try to reduce excessive leverage which increases the impact of speculation. This can be done through regulation for financial institutions by increasing reserve requirements, but I don’t know whether there would be other types of speculators exempt from such requirements.
Sam – I am somewhat confused by your argument that “welfare” improvements are compatible with slowdown in GDP growth. The lower gdp growth certainly wasn’t reallocating wealth from richer to poorer or improving average family incomes. Real median family income in the US, for one, went from growing fast in the pre-liberalisation period to almost flat growth post 1980:
Growth from 1950 to 1980: 2.3 percent per year
Growth from 1980 to 2007: 0.7 percent per year
http://krugman.blogs.nytimes.com/2009/11/07/reagan-reagan-reagan/
I’d suggest that if this is an “efficiency” gain, however narrowly defined, it is one we can do without. Nor do I see how this can be construed as a “welfare” gain.
Nor do I find it easy to believe that, given the multiple world financial crises triggered by speculative capital flows, that “noise” is not a prime mover of financial markets (the alternative being that crises were deliberately engineered). Krugman again:
“Since the early 1980s there have been three big waves of capital flows to developing countries.
The first wave was to Latin American countries that liberalized trade and opened their markets in the wake of the 80s debt crisis. This wave ended in grief, with the Mexican crisis of 1995 and the delayed Argentine crisis of 2002.
The second wave was to southeast Asian economies in the mid 90s, when the Asian economic miracle was all the rage. This wave ended in grief, with the crisis of 1997-8.
The third wave was to eastern European economies in the middle years of this decade. This wave is ending in grief as we speak.”
http://krugman.blogs.nytimes.com/2009/11/09/finance-mythbusting-third-world-edition/
Alternatively, if “noise” does not significantly destabilise markets as you claim, since a Tobin tax is essentially a tax on noise there is no reason to claim that it would significantly destabilise markets either.
For a specific example you could get your analytical teeth into, may I suggest Robert Pollin and Dean Baker’s paper: “Securities Transaction Taxes for U.S. Financial Markets”, http://www.peri.umass.edu/236/hash/aef97d8d65/publication/172/
I will make some general points which may or may not be ignorant:
1) Although it is technically true that all taxes, except those on economic rent, are “distorting” (again, begging the question of whether the market allocation is effective in the first place), the financial sector is effectively taxed much less than other sectors: for example stamp duty on real estate transfers is much higher than any tobin-type tax. This imposes a much greater distortion in that it encourages too much investment in the sector relative to the rest of the economy. Tobin tax revenue could be used to reduce other taxes which are far more distorting and welfare affecting, e.g. payroll tax.
2) As short-term investing in the financial sector is pretty much the perogative of the top 5% or less of the population (see ”Temporal evolution of the `thermal’ and `superthermal’ income classes in the USA during 1983-2001″ by A. C. Silva and V. M. Yakovenko
http://dx.doi.org/10.1209/epl/i2004-10330-3) (M&D owners of 2nd tranche telstra shares don’t count, except as an example of allocating risk to those who had no idea how to deal with it), tobin taxes progressively redistribute income, unlike other consumption and transaction taxes such as the GST which are regressive.
3) Advocates of a Tobin tax argue that it will decrease economic volatility. If true, then it would perform a valuable Pigouvian or externality tax function as it would force traders to pay for the volatility they generate, thereby reducing the risks of fluctuations etc against which banks and other long term investors are forced to hedge in the first place.
Cheers, James
<i>A large volume of trading is induced by intial hedging of the risk in spot and forward forex and IR markets</i>
“Spot” doesn’t come into it as all they have to do is a forward swap.
When did it become impossible to enter a forward exchange agreement?
An interesting take on Tobin taxes from the neo-Chartalist perspective (by the Research Professor of Economics at U. Newcastle) here:
http://bilbo.economicoutlook.net/blog/?p=5932
A key quote:
“Observation 1: The volume of financial transactions in the global economy is 73.5 times higher than nominal world GDP, in 1990 this ratio amounted to “only” 15.3. Spot transactions of stocks, bonds and foreign exchange have expanded roughly in tandem with nominal world GDP. Hence, the overall increase in financial trading is exclusively due to the spectacular boom of the derivatives markets …”
If the story Sam tells about the operation of markets being about facilitating global trade and efficient risk allocation is correct, one would expect
a) all sections of this market to expand in tandem, as opposed to a boom in one sector, a clear sign of a bubble.
b) that the efficiency of risk allocation would have expanded ~5 fold over the period 1990-2008. Let’s see, this period gave us the Latin American debt crises, the SE Asian financial crisis, the dot-com bubble, and now the Global Financial Crisis. Not really a sign of efficient allocation of risk.
Paul Krugman advocates a Tobin Tax:
http://www.nytimes.com/2009/11/27/opinion/27krugman.html
17December 2009. Kevin Rudd, wisely and in the interests of the Australian population, its business, self-funded retirees and financial institutions, has dismissed any possibility of a Tobin tax.
Fortunately for the beneficiaries of an adverse trickle-down effect, he stood up to bullying, less financially stable developed and un-developed leaderships. His clarification on the Tobin tax as not something that our country would be interested in, is the best thing to happen. Tobin himself, after several years contemplation, rejected the overall value of the tax he devised. It was in use (I think Sweden) but that country ultimate saw the wisdom in ridding itself of a highly inequitable, complex and failed instrument.
Australians should be very pleased that their Prime Minister, no matter what his faults, stood up to the bullies who were trying to drag him into gloabalisation of our independent Australia’s financial businesses.