Chief Economist, Lars Frisell delivers the 5th Paddy Ryan Memorial Lecture at Galway-Mayo Institute of Technology

11 December 2012 Speech

The economics and psychology of financial crises

I want to start by thanking the Galway Chamber of Commerce and the Galway-Mayo Institute of Technology for inviting me tonight and for the honour of delivering this year’s Paddy Ryan Memorial Lecture. I will share some thoughts on both economic and psychological aspects of financial crises. Most of these ideas are certainly not my own but of greater thinkers, and have been conceived of by behavioural scientists several decades ago, often in classroom experiments. The generations of psychology and economics undergraduates that have been subject to all sorts of behavioural experiments are worthy of gratitude, but I believe the experiments have been mostly harmless. I think the fundamental insights thus gained into the human psyche have a lot to say about why societies repeatedly experience financial crises, and why recovery from them tends to be so hard.

Let us start with the boom preceding a crisis. Now and then the news of a technological innovation, a new investment opportunity or just a general notion of better times ahead gain traction among investors. One’s future profits will be larger the more and the sooner one invests, so demand rises quickly and - with limited supply of the asset - prices soar. Keynes famously attributed this tendency for over-confidence to our “animal spirits” - and to our greed. The assets we tie our hopes to differ from crisis to crisis. They have included tulips, railways, internet start-ups, and, most commonly, property. According to one account, the price of one tulip bulb in Netherlands in 1637 equalled ten year’s annual income of a skilled craftsman.At its initial public offering in February 2000, the market capitalisation of, an internet retailer launched less than 18 months earlier, reached over one billion dollars. It was liquidated in November the same year after its stock price had fallen by 98 per cent. The most well-known manifestation of the Irish property bubble – or perhaps any property bubble - is probably the sale of the Burlington Hotel in Dublin in 2007 for some €288 million, or the equivalent of 600 000 euro per room. This is more than most luxury hotels in Manhattan or Dubai, both then and today.2

In retrospect it is of course blatantly clear that asset prices in all these episodes were driven far beyond their fundamental value. But is it possible that the valuations made at the time seemed reasonable? I have heard that a room-by-room comparison in the case of the Burlington is unfair, as Mr. McNamara apparently planned to add to this cost by building something much bigger. And one should perhaps concede that the demand for tulips in the 17th century, as well as for on-line pet clothes in the 21st, may have been easy to miscalculate. But still. Economists are usually very reluctant to recognise irrational behaviour as an explanatory factor for any sequence of events. After all, the basis of the economics discipline is the “economic man”, a profit-maximising agent who possesses unbounded cognitive ability. The dilemma economists face when they try to reconcile asset bubbles with rational behaviour is well illustrated by the famous late economic historian Charles P. Kindleberger. Kindleberger described asset bubbles as “[n]ew opportunities for profits are seized, and overdone, in ways so closely resembling irrationality as to constitute a mania.”In the 1990s Alan Greenspan went a bit further when he coined the term “irrational exuberance” to describe the rising stock markets of the 1990s.I think we should contend once and for all that asset prices in these episodes, and the expectations of many of the investors involved, were completely senseless.5

Two insights from psychology can help explaining how we end up in this anomalous state of affairs. The first is how we form price expectations. Far from the forward-looking behaviour of the economic man, price expectations are to a large extent formed through a backward-looking process. Hence, rather than becoming increasingly sceptical of an asset which (relative) price has inflated over recent years, we start considering it a normal development and expect the price to continue to increase. (The documented stickiness of inflation expectations, also in environments where radically different monetary policy regimes have been implemented, is a case in point.) Even more important, our sense of what is a reasonable price for an asset is embarrassingly flimsy, an effect psychologists refer to as anchoring. Suppose you have a pretty good idea of what you are willing to pay for a house, but just before you call the broker you hear that another house just sold for double that amount. It may be a very different kind of house, but that number sticks in your head and tends to increase your own bid. Marketers are well aware of this effect of course, and the sad truth is that the oldest trick in their book, “SALE! - previously y euros now only x”, actually works.

But overpriced assets in itself is not necessarily a matter of great concern. Many asset bubbles, including the IT-crash and Dutch tulipmania, had relatively small consequences for society at large. For sure, when asset prices rise and fall over a short time period there will be winners and losers, but it doesn’t mean there is a net loss. The reason why property bubbles tend to be so costly is that they bring about significant misallocation of resources. A certain number of the 90,000 or so housing units that were produced in Ireland at the peak of the bubble – three or four times the annual demand – will never be lived in. An even larger number will significantly deteriorate before they find a buyer. This is the real cost to society, a huge waste of productive resources.

In addition to the build-up of unrealistic price expectations I talked about before, investors in real estate are exposed to another risk, the so-called fallacy of compositions. The fallacy simply means that an action, such as an investment, may be individually rational on its own, but turns out to be a bad idea when many people are doing the same thing. This fallacy often shows up in everyday life, I suffered from it last time I decided to take the car into town when it was raining. Had I known about the mass of vehicles already lined up on the highway I would have surely reconsidered. The long lead times in construction makes the fallacy more likely as current prices do not signal the houses or hotels under production. Hence, overinvestment may go on for a long time.

Probably even more important is the huge capital requirements of property investments and therefore the usually high levels of leverage involved. Leverage amplifies boom-bust cycles in at least two ways. First, when banks decide to lend to house owners and property developers they naturally take the property as collateral and use the (assessed) value of the collateral as a basis. In a rising market the value of the collateral rises, enabling more credit. More credit in turn enables more investment into the property sector, which fuels prices even more. This feed-back loop is known to economists as the financial accelerator. Of course, when prices eventually fall the accelerator works in reverse. Falling property values triggers margin calls, which drains the sector of liquidity and forces fire sales, speeding up the implosion of prices.

Second, as is well known, leverage distorts incentives and increases risk-taking. In normal times this does not become a problem on the aggregate as most firms and house-owners invest with a long-term perspective. But in more speculative phases leverage can be the source of extremely wasteful undertakings. Betting in the casino is normally not a good business proposition, as you would usually have only 18 chances of 37 to win on red, counting the zero. But if 50% of the wager is borrowed money – and you enjoy limited liability – you only need a winning probability of one third to make it even money. Many business ventures undertaken towards the end of asset bubbles have the character of bad bets.

Bankers themselves, whose job it is to ensure the credit worthiness of their borrowers, are of course not immune to moral hazard. There are often very tangible short-term gains in terms of career prospects and cash bonuses - sometimes of colossal magnitudes – from extending credit, while potential credit losses lie hidden in the future, and will only to a small extent affect the individual banker. It is mostly impossible to say which of all poor credit decisions were the result of general over-confidence and which stemmed from narrow self-interest. But it is probably safe to say that both factors are in play.

Where does all this lead us? For the reasons above I think we are eternally doomed to experience cycles of booms and busts. But although bubbles may involve virtually any asset, large misallocations of resources usually take place when significant leverage is involved. My conclusion is that this necessitates central banks and supervisors to have an explicit macro-prudential mandate. Not because we are better at forecasting prices or assessing credit risk than others, but because we can combine information from all strands of the economy, including every lender. Our task is to distill these masses of information to detect potential imbalances, and, if publicising our assessment is not enough to stem the tide, we must take action on basis of it.

Let me now turn to the resolution of crises, and more specifically on the repair of balance sheets and fiscal consolidation. I shall not comment on last week’s Budget, suffice it to say that it cannot be an easy task to introduce a series of such austere budgets as those that Ireland has committed to.

As you may be aware, there is an intense debate among academics on how to tackle the current debt crises. Crudely put, the question is whether we should continue to consolidate, or if we actually could spend our way out of it. The answer depends on the size of the so-called fiscal multiplier. The fiscal multiplier is the ratio of a change in national income to the change in government spending. If the multiplier is larger than one it means that budget cuts will lead to an even bigger decline in GDP, so that the debt-to-ratio in fact increases. Likewise, increased government spending will lead the economy to grow more than one-to-one so the spending decreases the debt ratio. A lot of bang for the buck in other words. The multiplier has usually been considered to be somewhere between a half and one, but new findings in the IMF’s World Economic Outlook this year suggested that, in today’s environment of high unemployment and interest rates close to zero, it may be substantially higher than one.

For me the multiplier debate is slightly artificial. First, there is of course not “one” multiplier - different budgetary measures will have very different impacts on incentives and growth. For example, I have not heard anyone argue that the proposed mansion tax will have significant negative effects on our business climate or on the labour market. Second, there are poorer countries than ours that have managed to push through severe austerity measures and then quickly return to a positive growth path. An illustrious example in recent years is Latvia.

Again, insights from the behavioural sciences may offer more to the debate on austerity. The first I want to mention is known as the endowment effect – the fact that it is much harder to give up something you once possessed than to abstain from getting it in the first place. (I’m not sure whether this contradicts the old saying that it is better to have loved and lost than never to have loved at all.) There is a famous experiment that well illustrates this effect.College students were divided into two groups. Students in the first group were shown a mug with the college crest, and asked how much they were willing to pay for it. Students in the other group, on the other hand, were first given the mug and then asked what they would sell it for. The results were revealing. As mug-owners the students attached about twice the monetary value to identical mugs as did non-owners.

A second psychological force is the human desire for fairness. There are numerous experiments that highlight how powerful this is. In the so-called ultimatum game one person makes a take-it-or-leave-it offer to another person on how to split a given amount of money. If the offer is rejected both people get nothing. What kind of behaviour should we expect to see in this simple game? The only “economic” solution to the game is that the responder accepts any offer, since this is better than getting nothing, and the proposer logically offers a very big portion to him or herself. But that is not what we observe. Indeed, people can sacrifice considerable amounts in order to punish what they perceive as unfair behaviour. Low offers, say around 20% of the total, typically have about a 50% chance of being rejected.Hence, even if austerity measures are economically efficient they may get little support if they are perceived as unfair – even when this hurts the nay-sayers as well!

This leads me to my second, and concluding, message. Austerity is difficult. Psychological resistance will be greater than the pure economic impact would suggest. It is therefore hard to overstate the importance of social cohesion and a fair distribution of burdens when countries embark on fiscal consolidation. Ireland has so far managed the arduous process of austerity better than most countries, and our future success largely depends on this continuing to be the case.


1 The event has been popularised in the book Extraordinary Popular Delusions and the Madness of Crowds, written by British journalist Charles Mackay in 1841.

2 The four-star Burlington, one of Europe's biggest city centre hotels, was bought by a consortium backed by Bernard McNamara. In November 2012 the hotel was purchased by Blackstone Group for €67mn, less than a quarter of the price in 2007.

3 Manias, Panics, and Crashes: A History of Financial Crises, page 2.

4 Greenspan's comment was made on December 5, 1996 in his speech The Challenge of Central Banking in a Democratic Society.

5 One of the economists who early recognized the role of human emotions in asset booms and busts is Robert J. Shiller. A good reference is Animal Spirits: How Human Psychology Drives the Economy, and Why It Matters for Global Capitalism (2009) by Robert Shiller and George Akerlof.

6 See “Fairness and the Assumptions of Economics”, Journal of Business 59 (1986), by D. Kahneman, J.L. Knetsch and R. Thaler.

7 See, for example “An experimental analysis of ultimatum bargaining”, Journal of Economics Behaviour and Organization, 3 (1982), by W. Güth, R. Schmittberger and B. Schwarze.