The ideas, as published
Everything in this section is a paraphrase in this site’s own words. Nothing is quoted. The works it comes from are listed in the next section.
- Everyone’s view is partial.
- Soros starts from fallibility: the people who take part in a market understand it only in part, and their views are always biased in some way. He applies this to himself first.
- Reflexivity: the loop runs both ways.
- In the usual account, prices follow the facts. In The Alchemy of Finance prices also change the facts. A company whose share price rises can raise money cheaply, which improves its results, which seems to justify the price. Rising property prices make lenders more willing to lend, which raises prices further. The book works through the conglomerate boom of the 1960s and the property trusts of the early 1970s as cases.
- Boom and bust have a shape.
- His model starts with a real trend and a mistaken belief about it. The two strengthen each other. The belief is tested by events and survives, which strengthens it more. Eventually the gap between belief and reality becomes too wide to sustain; there is a moment of doubt, and then the same loop runs in reverse. He notes that the fall tends to be faster than the rise.
- A position is a hypothesis.
- He describes forming a thesis, investing in it, and then watching whether events confirm it. Part of The Alchemy of Finance is a diary, kept from August 1985 to November 1986, in which he records his reasoning and his decisions as they happened, changes of mind included.
- Finding the flaw is the useful part.
- In Soros on Soros he describes, in paraphrase, looking for the flaw in each of his own theses and being more at ease once he knows what it is. Being wrong is expected. Staying wrong is the error.
- It is not a science, and he says so.
- The word alchemy in the title is deliberate. He writes that reflexivity does not produce firm predictions of the kind natural science does: it is a way of looking at a situation, not a formula. In his 2013 paper he also records that economists for a long time did not take the theory seriously.
The sources
Named by title, author, publisher and year so that they can be found and checked. No links are given.
- The Alchemy of Finance: Reading the Mind of the MarketGeorge Soros. Simon & Schuster, 1987; later editions from John Wiley & Sons.Used here for: Reflexivity, the boom and bust model, the historical cases and the diary of 1985 and 1986.
- Soros on Soros: Staying Ahead of the CurveGeorge Soros, with Byron Wien and Krisztina Koenen. John Wiley & Sons, 1995.Used here for: His own account of the sterling position of 1992 and of how he handles being wrong.
- The Crisis of Global Capitalism: Open Society EndangeredGeorge Soros. PublicAffairs, 1998.Used here for: His account of the Russian crisis of August 1998, in which his funds lost heavily.
- The New Paradigm for Financial Markets: The Credit Crisis of 2008 and What It MeansGeorge Soros. PublicAffairs, 2008.Used here for: Reflexivity restated and applied to the credit boom.
- Fallibility, Reflexivity, and the Human Uncertainty PrincipleGeorge Soros. Journal of Economic Methodology, 2013.Used here for: The theory in its most careful form, with its limits.
Two documented examples: the dollar in 1985 and sterling in 1992
The diary in The Alchemy of Finance opens in August 1985 with a thesis that the dollar, which had risen for years, was due to fall. On 22 September 1985 the finance ministers of five large economies met at the Plaza Hotel in New York and agreed to bring the dollar down. The diary records that he already held positions that would gain from a weaker dollar, and that he added to them after the announcement instead of taking the gain. The same diary records the later stretches in which his views were wrong.
The second example is the one he is known for. The United Kingdom had joined the European Exchange Rate Mechanism in October 1990, which tied the pound to the German mark within a band. By 1992 Germany was holding interest rates high after reunification while the British economy was in recession and needed them lower. His thesis, as he later described it in Soros on Soros, was that the British government could not keep both the exchange rate and its economy, and would give up the exchange rate.
What made the position unusual was its shape. If he was wrong, the pound would stay inside its band and the loss would be small. If he was right, the pound would fall a long way. On 16 September 1992 the government raised interest rates twice in a day and then suspended sterling’s membership. The position has been widely reported at around ten billion dollars and the gain at about one billion, figures he has confirmed in interviews.
Two things are easy to miss. The thesis was about a fixed exchange rate defended by a government, which is a rare kind of situation: most markets offer no such one-sided bet. And the position was far larger than the fund’s own money, which is only possible with financing that a private account does not have.
When the thesis failed: 1998 and 2000
On 17 August 1998 Russia devalued the rouble and stopped payment on part of its government debt. Soros’s funds had large investments in Russia and lost heavily, which the firm acknowledged publicly at the time. He had written a letter to a newspaper a few days earlier arguing for a managed devaluation. He gives his own account of those weeks in The Crisis of Global Capitalism.
In the spring of 2000 the Quantum Fund lost heavily in technology shares as that market fell. At the end of April 2000 he announced that the fund would be reorganised to take less risk, and its senior managers left.
Both cases fit his own model, which is the uncomfortable part. Knowing that booms feed on themselves and then reverse does not tell anyone when. A theory that describes how a bubble works did not prevent its author’s fund from being caught in one.
The theory has critics on its own terms too. Economists have argued that an account which explains both a rise and a fall cannot easily be tested. Soros concedes much of this: he presents reflexivity as a framework for thinking, not as a means of prediction.
What can be studied with a small paper portfolio
What can be practised is the discipline of the hypothesis: saying in advance what would show a view to be wrong, and keeping the record honest. The exercise is done on paper.
- Write one view as a single sentence about a hypothetical position. Under it, write what would have to be true for the view to be right.
- Write what you would expect to see if it were wrong, in terms that can be checked, and a date on which you will look.
- On that date, record two separate things: whether the reason you gave came about, and what the price did. A price that moved your way for a different reason is not a confirmed thesis.
- Keep the entries in which you changed your mind, and note how long it took. That figure is the one this method is about.
- On the Rule bench, test any simple rule on one invented market and then look at its forty other results. It shows how often a result on a single run is chance.
- The journalA plan before and what happened after, for each entry. It stays in your browser.
- The Rule benchA rule tested on invented prices, where no rule has an edge, and forty other runs beside it.
A hypothesis written on paper costs nothing to hold. With money and leverage, the same hypothesis can be right in the end and still lose the account before then.
What does not scale down to a private account
- Financing terms
- The positions described were several times the size of the fund’s own money, financed by many banks on terms negotiated one by one. A private account has a fixed margin rule set by its provider.
- Derivatives access
- Selling a currency in that size is done through forward contracts and options arranged directly with banks. A private person cannot deal in that market.
- Liquidity, and size as part of the method
- A fund of that size is itself part of the pressure on a price: in a reflexive market, the selling helps to bring about what the seller expects. A small account’s order changes nothing. It gains only if others, for their own reasons, do the moving.
- Information and staff
- The theses were built with analysts, with long experience of how governments and central banks behave under strain, and with a wide circle of contacts. Reading the same newspapers is not the same position.
- Time horizon and staying power
- A large fund can hold a position through weeks in which it moves the wrong way. A leveraged private account is closed out when its margin runs short, so the view can be correct and the money still lost.
- Tax treatment and structure
- A fund and a private person are taxed differently, and the rules depend on the country.
- A leveraged CFD is not this
- A contract for difference on a currency or an index is a leveraged agreement with a provider, usually with a financing charge each night and a close-out level. It gives no ownership of anything. A view about an economy held through such a contract has costs and limits that the positions described here did not.
Questions people ask
- What is reflexivity in financial markets?
- It is George Soros’s term for a two-way loop. People’s beliefs move prices, and prices in turn change the underlying facts the beliefs are about: for example, a rising share price lets a company raise money cheaply, which improves its results. The loop can strengthen a trend far beyond what the facts first justified, and then run in reverse. It is set out in The Alchemy of Finance (1987).
- What did George Soros do on Black Wednesday?
- His fund held a very large position that would gain if the pound fell out of the European Exchange Rate Mechanism. On 16 September 1992 the British government raised interest rates twice and then suspended sterling’s membership, and the pound fell. The gain has been widely reported at about one billion dollars. His reasoning is described in Soros on Soros (1995).
- Was George Soros always right?
- No, and his own books say so. His funds lost heavily in Russia in August 1998, and in technology shares in the spring of 2000, after which the Quantum Fund was reorganised. His writing treats being wrong as normal and stresses recognising an error quickly. He also writes that reflexivity does not give firm predictions.
The words on this page
A summary for study of ideas that are on the public record, in this site’s own words. It is not endorsed by, or connected with, anyone named. Nobody at GIO4X has access to their portfolios. Studying a method does not reproduce its results, and a leveraged CFD is not ownership of an investment and does not behave like one. Educational information, not investment advice or a recommendation to trade.