Saturday, 4 January 2014

Avoiding bubbles with pooled funds

There's a simple formula which shows how a speculative bubble can grow. With present value (P) and a positive rate of return (r) to time (t) the value of the annuity (A) rises exponentially:

With a negative return the bubble tends towards zero. Since balance sheets have both assets and liabilities, there are four possible bubbles and anti-bubbles:

Financial Crisis Observatory (2013)

There are much richer mathematical possibilities to represent a return series, such as braid groups; the problem here is not so much the representation as the fact that there are bubbles. The uncertainty could be managed by simply pooling the tail risks, for assets and liabilities. The short-term characteristics are less susceptible to expectations, measurement errors and variance in the rate of return. Only as the future unravels do the long-term characteristics reveal themselves. The pooling of tail risks is a common insurance strategy and, when there is market failure, unravelling these long-term characteristics falls on governments and central banks.

Imagine, instead, if financial asset and liabilities were split into short-term and long-term annuities: the short-term being held privately, and the long-term being owned by a pooled or government fund. With an increase in the general level of risk, such as unmitigated climate change, the value of the long-term pooled fund would fall. With effective mitigation, the value of the long-term pooled fund would rise. If time (n) represents the length of the short-term annuity, the value of the long-term pooled fund to time (t) is approximately:

That gives the government or pooled fund a present value to assess risk mitigation projects. If the assets are firms or buildings at risk from climate change, the impact on the long-term value is what matters. Here, there is certainly scope for better maths and better estimates of the risks. Within this pooled fund we already have the tail risks for climate change, and assets that are absent from the national accounts: oceans, forests, air and water.

Whether or not private markets function at these longer maturities is highly unlikely. Poverty and high-risks are indistinguishable, because both are a preference for cash today. Long-term estimates for natural assets are heavily contested, such as the industry figures for carbon. There is no obvious private incentive to mitigate tail risks, except perhaps in a well-regulated and mandatory insurance market. Mitigation projects would convert an uncertain stream of insurance cash flows into a fixed sum: the higher value of the long-term annuity less the project cost. However, few (if any) firms and households insure events beyond next year, let alone beyond their lifetime.

So while pooling tail risks would seem to avoid the excesses of asset and liability bubbles, the quadruple problems of poverty, under-insurance, valuation errors and the government as an 'insurer of last resort' remain. There is no incentive for private interests to deal with tail risks. No-one will join a pooled fund today when they think they have the winning lottery ticket tomorrow. The politics will fail as long as the majority believe they are one step ahead of environmental and financial risks.

Financial Crisis Observatory. (2013). Financial Bubble Experiment. Available at:

Saturday, 7 December 2013

How to start a crypto-bank run

First published 7th December 2013
Updated 10th February 2014

A friend who is a crypto-currency evangelist gave me a Bitcoin a few years ago, which I jokingly accepted as a 'store of value' until the price bubbled* to over $1000. When I heard that China was banning direct trade in Bitcoins, I sold up and suggested my friend did, too. Yes, the flash-crash was China's fault. For a while I was tempted to roll the crypto-dice, buying Litecoins in expectation that Mt.Gox would start trading them; or arbitraging the exchanges**.

But I find it difficult to believe there is a 'natural price' or that 'par' will emerge from the chaos: I think insiders will profit and outsiders will lose. To exchange my Bitcoin I paid FOUR sets of fees: a percentage to the trading platform; a fixed spread to the settlement platform; an exchange rate fee from USD to GBP; and a transfer fee to a 'recognised' UK holder of GBP: total fees were over 10%.

Crypto-currencies have seen the most unexpected people jump into bed together: drug dealers, geeks, crypto-anarchists and libertarians. On the left, their anger is directed at private interests; on the right, it is directed at government.  But crypto-currencies look like Wildcat banking to me, with an unregulated network of international crypto-banks. There are multiple currencies and trading platforms, and you can buy derivativeslend and borrow.

Without anyone to uphold 'par' or regulate the crypto-banks the most obvious lesson is to have stronger regulation of the exchanges and crypto-banks. Today (10th February 2014) there was a flash crash on the BTC-e exchange where a Bitcoin was briefly worth $102. I think I had a lucky escape.


* This post on Bitcoins makes a good point about their self-reinforcing nature. The basic foodstuff of crypto-banks is the Bitcoin. If derivatives are settled in Bitcoin, and crypto-banks and firms are bought and sold in Bitcoins, their scarcity drives up the price and perpetuates the bubble. There are similarities here with GBP and USD, which might become less scarce if more trade were settled in other currencies.

** There is also an analogy with the 'carry trade'. For long periods there have been profits to be made if you buy on the Eastern Europe and Chinese exchanges and sell on MtGox (USD). This works best for insiders who can avoid bank fees (7% on PayPal).

Wednesday, 20 November 2013

Are there policy alternatives to Ireland's austerity?

This is a response to the PKSG presentation by Stephen Kinsella on 19th November 2013.

In essence, the stock-flow-consistent approach is ‘to a very large extent an exercise in accountancy – or perhaps logic is a more congenial word’ (Godley, 1983). The paper by Stephen Kinsella extends this approach with a five sector model, estimated using OLS. Applying different spending and tax shocks, he concludes that Ireland's austerity could have been less severe.

Leading into the crisis, Ireland was running small government and current account surpluses. As expected under the 'New Cambridge hypothesis', the private sector ran a deficit. Unfortunately, this was due to investment in property rather than production. Remarkably, the assets and liabilities of the banking sector doubled between 2005 and 2008. 

Then, the Irish economy saw major shifts in the composition and direction of capital flows after 2008. Households became net savers, the foreign sector moved deposits our of Ireland and shifted to investment in equity. These effects are dwarfed, however, by government which, in the form of an IMF bridging loan, bailed out the financial sector (purple and red):

The 'run on deposits' followed the sale of Merrill Lynch to the Bank of America, on the same weekend in 2008 that Lehman Brothers collapsed. If the Federal Reserve had not prevented the collapse of AIG and Merrill had been left to fail, the story would be different: Bank of America is still being pursued by AIG for alleged fraud at Merrill. With the subsequent deleveraging, Irish banks are reporting costs higher than income and Merill looks the most fragile:

So while the paper recommends less austerity, the banking sector remains a risk for the Irish government. Yet fiscal policy is hemmed in by budget rules from the European Commission. The Treat of Maastricht requires that the government runs a deficit of no more than -3%, while the European Commission Alert Mechanism Report asks for a current account balance between -4% and +6%. Ireland has missed these targets every year since 2006:

The trajectory, above, suggests some uncertainty that Ireland will reach the green (private sector surplus) or red (private sector deficit) triangles (government deficit less than 3%, and a current account balance between -4% and +6%).

While the SFC model in the paper fits the data very well, like any SFC model it might be further disaggregated: in particular the treatment of shadow banking (ie: leverage); the derivatives book and the effects of currency and duration mismatch. Of course, such data are not readily available. Household spending might also be impacted by wealth, with realised and unrealised capital gains affecting behaviour to varying degrees: Minsky's hedging, speculative and Ponzi modes of financing. Ireland's low corporation taxes also encourage profit retention via foreign investment.

There are other policy alternatives for Ireland: having symmetrical rules for government budgets at the EU;allowing fiscal transfers from 'core' to 'periphery'; and similar mechanisms such that surplus countries pay towards adjustment costs. So as well as the risks from a weak household sector under austerity, there is a risk that foreign capital will dry up or reverse due to tax treaties or tax harmonization being imposed at the EU level. Not that there are many calls for this in Ireland. As Stephen put it succinctly: 'turkeys don't vote for Christmas'.

Wynne Godley (1983), Personal correspondence with J.K. Galbraith

Tuesday, 12 November 2013

India's 'liquidity call'

What is the Reserve Bank of India up to?  On Perry Mehrlings MOOC on the 'Economics of Money and Banking', the students are (quite rightly) asking why interest rates were rising. So while the Fed, ECB, Band of Japan and Bank of England continue with their 'liquidity put' to the international banking system, the Reserve Bank of India is doing the opposite: a 'liquidity call'. I'm hopeful that Perry might weave a few 'liquidity calls' into his own discussions of high interest rate economies.

As the next graph shows, it's not as if inflation was rising prior to the rate hikes in February 2012. Rather, the hike was a response to the fall in the Rupee that began, in earnest, around August 2011 when inflation was actually falling:

There are indications that the Reserve Bank of India was fixating on inflation targeting in a speech by Dr. Subir Gokarn, a Deputy Governor in June 2011: 'ultimately, inflation outcomes will determine the monetary stance' (Gokarn, 2011, p. 2). As soon as the 'liquidity call' was made, there were profits to those who could second guess the timing and impact.

Another speech in February of that year set the scene for foreign investors. The Reserve Bank of India saw 'gross capital flows contribut(ing) immensely to diffusion of technology and international knowledge flows' (Gopinath, 2011, p. 1) and was keen to avoid 'the "original sin" of excessive foreign currency borrowings by domestic entities' (Gopinath, 2011, p. 6). In other words, India was clamping down on domestic entities borrowing abroad, opening up to foreign institutional investment and seeing the way forward as 'continuing liberalisation' (Gopinath, 2011, p. 14).

The problem is that inflation did not fall. Rather, asset prices rose. The Indian stock market 'hit a record high, propelled by an increased inflow of foreign capital' (BBC, 2013). The rather stubborn inflation suggests  that firms are able to mark up their prices and pass on their financing costs to consumers' along with higher commodity costs. That sounds more like a supply problem.

One of the business school cases we teach is that India needs more investment, not chasing asset prices but investing in the supply chain. Gopanith is right that the world benefits from the flow of international knowledge and technology, but must this must go hand-in-hand with financial flows?  The 'cold chain' case that we teach was researched at the India Development Foundation in Mumbai and the Indian Institute of Technology in Delhi: domestically-produced knowledge. These researchers find that 'international trade could be a powerful engine of agribusiness growth in the future.... India could be a sourcing hub for products such as rice, organic produce and food products such as ready to eat meals' (Khan, 2005, p. 13) if only there were investment in logistics, transport and refrigeration. As it stands 'around 35 percent to 40 percent of the total production of fresh fruits and vegetables, is wasted in India, which is about the total production of the Great Britain' (Joshi et al, 2013, p. 1263).

This is a problems with under-investment.  Joshi et al. suggest several solutions: an industrial policy that supports industry players to upgrade infrastructure; the domestic use of refrigerated shipping containers; a reduction in excise duty on processed food; better internet and mobile connections; and greater assurance about the safety and quality of food.

By raising interest rates, the central bank is stifling real investment at a critical moment for India.

BBC (2013),Indian stock market hits record high. Available at
Gokarn, S. (2011), Striking the Balance between Growth and Inflation in India, CII and Brookings Institution, Washington DC. Available at
Gopinath, S. (2011), Approach to Capital Account Management - Shifting Contours, Annual Conference of the Foreign Exchange Dealers’ Association of India (FEDAI), Thimpu. Available at:
Joshi, R., Banwet, D.K. and Shankar, R. (2009), Indian cold chain: modeling the inhibitors, British Food Journal, 111 (11),  pp. 1260 - 1283. Available at
Khan, A.U. (2005), The domestic food market: is India ready for food processing?, Conference
Proceedings, Pune: India. Available at:

Monday, 4 November 2013

Mathematics for New Economic Thinking

Last week, INET and the Fields Institute brought together economists and mathematicians in Toronto. The results were certainly interesting... with several economists paying their homage to Wynne Godley, including Stephanie Kelton, Marc Lavoie, Peter Skott and Randall Wray. It was interesting to see these economists play the same stage. Some speakers felt that bad economics had tainted mathematics: fewer believed the reverse.

I am drawn to Wynne Godley’s models: at least, those versions influenced by Minsky. Naively, perhaps, I was hoping that a new paradigm would emerge from the ashes of the 2008 crisis. Far from it... put two economists together and you get more than two theories... like politics, economics is messy and confrontational. At this conference, the data were sometimes assumed and the theories were more often debated.

There were some contrarians. Perry Mehrling made an eloquent plea for better data on the stocks and flows, something that would surely reinvigorate deductive and descriptive economics. Alan Kirman called for models that explain crises and inequality, rather than such awkward complications being exogenous. The questions were direct: what if the far right interpret MMT as a reason for states to abandon the Euro or, even, their credible commitment? As I write this, I wonder about the symmetry to Perry’s liquidity put at the Fed: countries, like India, where the central bank has placed a liquidity call and speculators are making ‘carry trade’ profits.

There was even a model that confirmed Keynes’ insight that 'there is no limit to the amount of bank-money which the banks can safely create, provided that they move forward in step' (John Maynard Keynes, 1930) . Sometimes, this emphasis on banking left me wondering if the model had lost sight of the instability… with banking assets taking precedence in crises and equity holders absorbing the losses. Stability is not destabilizing, if you have a central bank behind you.

Of the more empirical papers, Didier Sornette gave an excellent presentation that showed few bubbles are preceded by volatility. The relatively simple maths suggested that leasehold assets have a safety valve built-in. Yet the politics are very different if the leaseholder is the Duke of Westminster, or the Chinese government. Either way, if you were to 'deflate the bubble’ by playing with the terms of the lease would you simply turn a financial bubble into political turmoil?

Sadly, not every presenter showed their love of math as simply as Edward Frenkel had the week before. Some of the pleas seemed unnecessary... why would I want to calculate inflation using dynamic field theory? My love, and hatred, of math is because numbers are also social... we irrationally fear Friday 13th, and in the UK we stick doggedly to the penny long after it has any monetary value.

As Matheus Grasselli told me, so eloquently, the Plano Real was essentially social and synthetic, but was able to calm hyperinflation in Brazil by giving an alternative unit of account. Sometimes the simplest maths can improve lives.

Tuesday, 30 July 2013

Should Barclays borrow more without conditions?

Would you unconditionally invest billions more in Barclays, knowing your equity was at risk and your loan would not be re-paid if the bank got into trouble? Since you get all of the downside, what return would you be looking for? 5%? 8%? Would you want the 14% that was given to Abu Dhabi and Qatar investors in 2008... a deal that is apparently under investigation by both the Serious Fraud Office and the Financial Conduct Authority? Or would you ask for the kind of conditions that the rest of the country has to accept, in particular pay restraint?

I should declare my link to Barclays: I worked at Barclays Global Investors (BGI) until 2007. Then, we were doubling in size every year and the culture had been risk averse. After all, we were playing with someone else's money.

But my last project was different: it had Bob Diamond and 'casino banking' written all over it. We were going to replace the tiny bit of cash in your pension with 'synthetic cash' that was made up of unintelligible, but AAA, bonds. When I questioned it, I was told that everything was going to be fine and this really did matter more than anything else. We were told to do similar things with 'synthetic ETFs' backed, not with real assets, but a promise to pay from an investment bank.

So I resigned. To be fair, it was messy and we spent some time seeing if we could sort things out, but like any failed marriage neither of us was going to change. A former colleague told me those unintelligible AAA bonds lost BGI around £200 million and, after a bit of soul searching, they lost their enthusiasm for 'synthetic ETFs'. By then, Bob Diamond had sold BGI for £8.2 billion. The jewel in the Barclays crown, so to speak, had been sold.

That left Barclays dominated even more by the investment bank. You might thing that £8.2 billion from Blackrock and £4.75 billion from Qatar would plug a pretty big hole in the balance sheet. Yet the annual bonus pool is up to £2 billion and there has been one provision after another: £1.5 billion for mis-selling interest rate derivatives, £2.6 billion for mis-selling PPI and £290 million fine for manipulating LIBOR.

So would you lend more money to Barclays? If they were a relative, and every time you lent them money they spent it, wouldn't you be a little bit more strict? OK, so Bob Diamond has left but is it really sensible to lend to the current management team without any conditions?

There's a double moral hazard in banking. First, the central bank steps in as 'lender of last resort'. Second, the industry turns to itself for funding via the capital markets. Yet there's a real economy outside London that needs investment but is constantly being told to minimise pay rises, accept less favourable employment contracts, find new markets and learn new skills. With other bankers and investment managers making the decisions, will those same conditions apply to Barclays?