Bubbles in Asset Markets - A critical valuation of experimental studies

Seminar Paper, 2012

16 Pages, Grade: 1,00



1. Introduction

2. A Definition of bubbles

3. Why do bubbles emerge?
3.1. Experience of Traders
3.2. Riding the Bubble
3.3. Short Selling Restrictions
3.4. Agency Problems

4. Are experiments suitable for testing the development of bubbles?
4.1. Knowledge of the Fundamental Value
4.2. Payout Structure

5. An Alternative to Experiments

6. A New Design for Experiments
6.1. Definition of the Fundamental Value
6.2. Identifying the Purpose of Trading
6.3. Short Selling restrictions
6.4. Payout structure

7. Conclusion

8. References

List of Abbreviations

illustration not visible in this excerpt

1. Introduction

Probably everybody knows at least one good example of a bubble in the past, like the recent real estate bubble in the United States that bursted in 2007. In hindsight most bubbles are stated as predictable, but nevertheless it happened several times and will happen again. Several experiments have been conducted to find reasons for the development of bubbles. But not all results seem to be very satisfying. The purpose of this seminar paper is to point out the shortcomings of past experiments and to suggest an alternative design for experiments which results in more representative findings. It further should give an overview about the most prominent explanations for the formation of bubbles and an evaluation of them. Thereby, the explanations in this paper are based on accepted financial papers and experiments of Harrison and Kreps (1978), Smith et al. (1988), Garber (1990), Allen et al. (1993), Lei et al. (2001), Noussair et al. (2001), Dufwenberg et al. (2005), Haruvy and Noussair (2006), Kirchler et al. (2011), and Xiong et al. (2011). Most of these papers conduct experiments with different research questions. Some are taken as a theoretical basis for the argumentation in this paper.

First I will give a short definition of bubbles in chapter 2 and quote the most prominent explanations as reasons for the emergence of bubbles in chapter 3. Afterwards I will point out some problems of the usual experimental design (chapter 4), shortly mention an alternative method to experiments (chapter 5) and finally propose a new experiment design in chapter 6.

2. A Definition of bubbles

Bubbles in asset markets are continuous increasing prices far above its fundamental value (FV), followed by a crash, the burst of the bubble. Since the FV of an asset is usually not exactly observable some deviation from that “true value” does always exist. Thus, there have been thousand of small bubbles in the past, but normally they are not called bubble. This name is only given to periods of undeniable overvaluation which are too far away from its FV and resulting in a huge crash. Garber (1990) names three famous bubbles: the Tulipmania in the 1630s, the South Sea Bubble and the Mississippi Bubble both in 1720. All of these examples show price drops down to 10 percent and less of their peak value.

Some economists would also name other, more recent bubbles like the dot-com bubble or the real estate bubble. Bubbles itself are not very harmful. The burst of bubbles is the real problem. Bubbles which never burst and smoothly approach toward its FV are nothing damaging. Actually the loss in value can be equal, but it is spread over a longer period and mostly over more investors, thus the “perceived damage” seems to be less. Of course, if considering undesired distribution of investments into assets which are currently in a bubble, this would lead to higher losses. The investment in strongly overvalued assets reduces the availability of funds for other assets and thereby reducing the overall value of assets. An overreaction of the burst could also lead to the bankruptcy of some firms or assets which also destroys value. Furthermore the burst of bubbles usually increases uncertainty in markets and therefore some investments may be cancelled. All these aspects are clear arguments for the harmfulness of bubbles.

3. Why do bubbles emerge?

Generally spoken, bubbles emerge because the majority of market participants strongly overestimate the FV of assets up to a certain point or event, most likely the appearance of information which dramatically changes their opinion, leading to a crash of prices. As already mentioned the FVs are unknown, but they can be estimated quite accurate at least by investors having the required abilities to do so. But how can these large estimation errors arise?

This chapter will discuss some crucial conditions leading to the development of bubbles.

3.1. Experience of Traders

Dufwenberg et al. (2005) examine the influence of experience of traders on the creation of bubbles. This is done by experimental research. They create a market were assets are traded with a finite life time of ten periods and a dividend payment of either 0 or 20 in each period. Thus, the FV defined by the future dividend payment is publicly known and decreasing. In total four rounds with different experience levels of traders are conducted. Their main findings are that the price deviation from the FV and the amplitude of prices decrease with traders becoming more experienced. Thus, experienced traders highly reduce bubbles in asset markets. But it is questionable if these experimental results are also true for real markets where the FV is unknown and usually gently increasing.

The FVs in these experiments do not change between rounds and thus there must be a positive effect of experience, simply caused by a better understanding of the experiment mechanism or the calculation procedure. In real markets with changing prices, values, participants and so forth experience could also have a negative value because some might falsely believe that past events will happen again. Regarding these considerations the experience of traders seems not to be a sufficient explanation for the development of financial bubbles.

3.2. Riding the Bubble

Since financial markets are a zero-sum-game bubbles are also resulting in profits for some investors. The burst of a bubble only retracts the prior gains some investors made. Thus, riding a bubble may be a reasonable strategy if ones assuming a further increase of prices. But the assumption of increasing prices is crucial. Some investors may overestimate the asset and hence they invest or some may believe that most of the other traders will overestimate future prices and try to participate financially by also investing. However, this requires the assumption that traders do not exactly know the FV, which would otherwise rule out heterogeneous believes. Without these uncertainties in valuation, called noise by Black (1985), trading in financial markets would be far more illiquid and prices would be close to the FV. Concluding, riding the bubble may be an appropriate explanation for real market bubbles, but not for experimental with known and decreasing FV. In experimental markets one could expect the asset price to be the same in the next period, even if the FV is actually decreasing and thus invest. He would make profit threw that investment if he gets a dividend payment in that period and sell it afterword at a price which is more than the buying price minus the dividends. This would be a reasonable strategy. But the cause of that strategy actually is that some participants do not know (or understand) the FV and thus drove the price upward.

3.3. Short Selling Restrictions

Short selling restrictions in experimental markets limit the amount of shares which can be sold by participants. Without those restrictions the supply of shares is much higher, whereas the demand is less affected. Haruvy and Noussair (2006) tested the impact of short selling restrictions on prices in experimental markets.


Excerpt out of 16 pages


Bubbles in Asset Markets - A critical valuation of experimental studies
University of Innsbruck
Catalog Number
ISBN (eBook)
ISBN (Book)
File size
578 KB
bubbles, asset, markets
Quote paper
Daniel Hosp (Author), 2012, Bubbles in Asset Markets - A critical valuation of experimental studies, Munich, GRIN Verlag, https://www.grin.com/document/198948


  • No comments yet.
Read the ebook
Title: Bubbles in Asset Markets - A critical valuation of experimental studies

Upload papers

Your term paper / thesis:

- Publication as eBook and book
- High royalties for the sales
- Completely free - with ISBN
- It only takes five minutes
- Every paper finds readers

Publish now - it's free