Economic Bubbles: A Story of New Eras, Emotional Contagion and Structural Support
©2014
Textbook
84 Pages
Summary
Economic bubbles go hand in hand with financial inventions, financial liberalization and excess leverage. Frequently these bubbles are fueled by the overoptimistic outlook not only of the so-called experts or gurus but also by the extremely positive perception of the general public. Great hikes in asset and commodity markets are believed to be a result of the new economy that has been created. Historical levels of markets and where the level of fundamentals should really be are either unknown or completely ignored. Partially one could account this to the short financial memory of market participants. Partially the abstract nature of the markets and the complexity of financial products themselves may be the reason. This paper tries to identify regularities defining economic bubbles concentrating on the most recent ones. Different approaches in explaining market movements will be discussed. Further, not only structural but especially psychological factors which may cause the emergence, the inflation and the implosion of economic bubbles are considered. Already existing agent based models, policy responses and possible future policy measures are analyzed and evaluated.
Excerpt
Table Of Contents
8
psychology and the inherit feedback mechanisms must play a major role
when prices surge and excitement about economy, technology, houses or
great investment opportunities spreads through social ranks, even
crossing borders infecting other markets at high speed. This study tries to
detect circumstances and components that repeatedly cause markets to
slide towards extremes. Particularly the influence of psychological factors
will be considered.
The paper is structured as followed:
Section two gives a brief overview of the Japanese asset price bubble, the
dot-com bubble as well as the subprime crisis. Section three will cover
theoretical approaches in explaining market movements, discussing the
appropriateness of the efficient market hypothesis, statistical approaches
and agent based financial market models. Section four and five consider
structural and psychological factors leading to the emergence, propelling
and the burst of speculative bubbles. Throughout section six policy
measures will be analyzed. Section seven concludes.
9
2.
The course of a typical speculative episode
Speculation: "The act of trading in an asset, or conducting a financial transaction, that
has a significant risk of losing most or all of the initial outlay, in expectation of a
substantial gain."
4
Speculative episodes seem to be a reoccurring phenomenon not only in
financial markets but also in real estate and commodity markets.
Historically, asset price bubbles in industrialized countries are somewhat
of a rare occurrence. However, there have been at least four distinct asset
price bubbles in the recent past starting with the Tokyo asset price bubble
in the 1980s, the real estate and stock price bubble in the Nordic countries
of Europe at around the same time, the Asian financial crisis in the 1990s
and the dot-com bubble followed by the subprime crisis in the late 1990s
and early 2000s.
5
Not only did these bubbles occur in a relatively short
time span, they also involved both the stock and the real estate market.
Keeping this in mind a short description of the two bubbles with the
greatest momentum will be given.
2.1
Japan and the Lost Decade
After the devastations of World War II, Japan made great technological
progress during the 1950s and 1960s. By the 1980s it was the second
leading industrial power dominating the global electronics industry.
6
The
Japanese market was highly regulated and protected by the government.
Entering the promising Japanese market was extremely hard and
sometimes even impossible for foreign companies. The Ministry of Finance
maintained low interest rate ceilings for deposits and loans and
government officials identified Japanese companies that were to be
preferred when it came to granting these low-interest credits. As inflation
was greater than interest rates on deposits, investment seemed to be
limited to real estate and stock, where real returns had been positive.
4
See: http://www.investopedia.com , retrieved 05/13/13.
5
See: for example Kindleberger, C. P. / Aliber, R. (2005), Akerlof, G.A. /Shiller, R. J. (2010), or
Reinhart, C.M. / Rogoff, K.S. (2011).
6
See: Kindleberger, C. P./ Aliber, R., "Manias, Panics and Crashes" 5th edition, (2005), p. 145 ff..
10
With greater economic prosperity demand for housing as investment
vehicle rose resulting in an unmistakable upward trend in the value for
homes. Low interest rates and increasing international demand for
Japanese high tech products caused business investment to surge, stock
prices increased rapidly and the Japanese yen appreciated. Kindleberger
and Aliber (2005) identify three factors that ultimately led to the
emergence of the Japanese real estate bubble. Firstly, the real (inflation
corrected) return on Japanese real estate had been positive for thirty
years making it look great as investment when compared to deposits,
where real returns, as already mentioned, where pushed into the negative
by higher inflation. Secondly, international pressure during the mid 1980s
led to deregulations of the Japanese markets resulting in its financial
liberalization. Interest rate ceilings on deposits and loans were raised,
governmental guidance was less extensive, restrictions on foreign
investment for Japanese firms were relaxed and banks were permitted to
increase their foreign branches.
7
Furthermore, banks became free to
increase real estate loans as share of their total loans. And thirdly, to
avoid further appreciation of the yen in order to maintain a competitive
advantage, the Bank of Japan opted for expansionary monetary policy,
increasing the money supply and further lowering interest rates. As
money supply increased, loans for real estate became relatively cheaper.
The optimism about the future and the feeling of financial superiority as
described by Galbraith (1994) in combination with high returns led to a
seemingly unstoppable inflow of funds into Japan. With increasing real
estate prices and a soaring economy many Japanese felt richer, trying to
diversify their new wealth turning to stocks. A second argument for the
increased investments into the stock market may be that while prices for
homes increased at a rate of 30% p.a.8 they became exorbitant for many
Japanese who, looking for investment opportunities started to invest into
the stock market.
Economic growth remained rapid during the 1980s, heightened
7
See: Kindleberger, C. P./ Aliber, R., "Manias, Panics and Crashes" 5th edition, (2005), p. 149.
8
See: Kindleberger, C. P./ Aliber, R., "Manias, Panics and Crashes" 5th edition, (2005), p. 152.
11
international demand for Japanese stocks, decreasing lending standards
and the anticipation of ever increasing real estate and stock prices with
possibly huge returns led to the inflation of both markets. Japan's stock
prices increased in real (inflation corrected) terms by 275.6% from 1982
to 19879 while the savings rate declined. Market participants seemed to
not evaluate historical price movements and did instead concentrate on
recent developments of the prices, extrapolating the trend. This behavior
of trend extrapolation in the short term is supported by the findings of
Shinji Takagi
10
who, through a survey among investors found, that
expectations in the short run are formed based on recent price
movements supporting the prevailing trend. Traditionally, Japanese banks
owned large amounts of real estate and stock. With increasing prices in
both, the real estate and stock market, the banks' capital also increased,
giving them the opportunity to further increase real estate loans. New
financial intermediaries were established to meet the great demand for
loans. Financial innovations in the form of convertible bonds emerged.
11
Formerly not eligible borrowers were now accepted and granted loans as
their perceived wealth in the real estate or stock market constantly
increased while lending standards at the same time constantly decreased.
With increasing stock prices firms could raise needed funds more easily
and at very little cost. Real estate prices increased faster than rents.
Initially investors wanted to meet the interest payments for highly
leveraged real estate through the expected rents. As rental income
dropped below the costs for the interest payments, the needed cash could
be made available through further borrowing against the real estate or
through selling the property. However, this option of finance was only
available as long as real estate prices increased at a rapid rate. As the
bubble grew, Japanese banks became some of the biggest banks in the
world, Japanese firms acquired companies and prestigious properties in
Europe and the US, the consumption of luxury goods surged and the
9
See: Schiller, R. J., "Irrational Exuberance" 2nd edition, (2005), p.141.
10
See Takagi, S., "Exchange Rate Expectations: A Survey of Survey Studies", (1991).
11
See: Kindleberger, C. P./ Aliber, R., "Manias, Panics and Crashes" 5th edition, (2005), p.152.
12
inflow of capital from Europe and the US seemed endless. "..The Japanese
had all the money and they were spending it ... as if they were very
rich..."
12
. By the end of 1989 home prices were so high that "...Banks had
developed one hundred year, three generation mortgages."
13
The
Nikkei index peaked at 38,916 in December 1989.
14
The market value of
Japanese stocks was twice the value of all US stock, the GDP of Japan
however was less than half of the US GDP. Land per capita in Japan was
four times as much as land per capita in the US, but income per capita in
Japan was only 60% - 70% of that of the US.
15
As the incoming governor
of the Bank of Japan worried about the social atmosphere in the country,
new banking regulations were brought on the way. The monetary policy
was tightened which raised the discount rate from 2.5% in May 1989 to
6% in August 1990 aiming at the stabilization of the markets and the
yen.
16
Japanese banks were instructed to limit the growth rate of real
estate loans as compared to other loans. Rental income however was still
not sufficiently large for the owners to meet the monthly mortgage
payments. And with a slowing growth rate of real estate loans the cash
needed to meet these obligations could no longer be obtained. Investors
became distressed sellers of real estate and stock causing the prices to
drop. Stock and real estate prices started to decline in January 1990.
Investors looking for more promising opportunities began to transfer their
funds away from Japan to other Asian countries as well as to the US.
Kindleberger and Aliber (2005) argue that this shift of funds towards the
Asian countries and towards the US stock market may have been the
foundation for the following speculative episodes in these countries.
Stock
prices declined by 30% in 1990 and another 30% in 1991.
17
They hit
bottom in February 2009 with the Nikkei being at 7,568, although the
global crisis at that time may also have had a negative impact.
18
A decline
12
See: Kindleberger, C. P./ Aliber, R., "Manias, Panics and Crashes" 5th edition, (2005), p.147.
13
See: Kindleberger, C. P./ Aliber, R., "Manias, Panics and Crashes" 5th edition, (2005), p.153.
14
See: http://finance.yahoo.com, retrieved 05/14/2013.
15
See: Kindleberger, C. P./ Aliber, R., "Manias, Panics and Crashes" 5th edition, (2005), p.146.
16
See: Schiller, R. J., "Irrational Exuberance" 2nd edition, (2005), p.224.
17
See: Kindleberger, C. P./ Aliber, R., "Manias, Panics and Crashes" 5th edition, (2005),p.154.
18
See: http://finance.yahoo.com, retrieved 05/14/2013.
13
in stock and real estate prices also meant that the Japanese bank capital
was declining at a fast pace. Lower bank capital and the new regulations
had a tight grip on banks which were more and more reluctant to lend.
Investment declined not only as a result of increasing interest rates and
changed attitudes of banks regarding the availability of loans. Anticipated
profits had also been revised downward while producing companies had
significant excess capacity. As people became worried about the future,
household spending slowed and increased at a much smaller rate than
formerly expected. The perceived wealth decreased and, to be able to
meet the interest payments, the savings rate increased and peaked in
January 1991 at about 34% of GDP.
19
Bankruptcies increased and so did
the loan losses for the banks and non-bank institutions putting them
further to the edge. Banks became increasingly cautious, what once was
thought to be conservative accepting real estate as collateral for loans
had become a highly risky endeavor. Internationally concerns about the
solvency of Japanese banks increased resulting in increased interest rates
for Japanese banks. The immediate impact was that many loans made by
offshore Japanese banks were no longer profitable and were called in.
20
Moreover, the increased premium for Japanese banks acted as a warning
signal for sensitive investors who further withdrew and re allocated their
funds to non Japanese banks, causing increased outflows of capital from
Japan. With the recession of 1991 Japan entered the era of what became
known as the Lost Decade where despite extremely low interest rates,
below 1%, the economic growth was lavish, oscillating around 0%.
21
2.2
Irrational exuberance in the US
The US stock market had been increasing since it hit bottom in 1982. The
first cell phone system was developed at that time with the use of cell
phones growing exponentially over the years. The World Wide Web
became available to the public in 1994. Both innovations had a major
19
See: http://www.tradingeconomics.com, retrieved 05/14/2013.
20
See: Kindleberger, C. P./ Aliber, R., "Manias, Panics and Crashes" 5th edition, (2005), p.153 ff..
21
See: http://www.tradingeconomics.com, retrieved 05/14/2013.
14
impact on how people were able to transmit and store information. They
changed the everyday lives of millions of people dramatically. The great
and fast advancements of technology could be experienced firsthand by
everyone.
22
In this atmosphere of a new technological and live changing
era, the stock market began its spectacular hike which ended in 2000
merging into a housing boom that ended with the subprime crisis in 2007.
Impressive US corporate earnings growth, 36% in 1994, 8% in 1995 and
10% in 1996
23
, coincided with the emergence of the new technologies and
were linked to them by the public. With a fast growing IT (internet
technology) industry, coupled with a weak dollar in the early 1990s
international demand for US capital and technological products increased
rapidly. The US dollar increased which made foreign goods relatively
cheaper for the US. The US economy boomed during the 1990 with an
inflation rate declining from 5.2% in January 1990 to 1.67% in January
1999
24
, the unemployment rate declined from 8% at the beginning of the
1990s to less than 4 % at the end of the 1990. The market value of US
stock increased to 300% of US GDP in 1999.
25
The personal saving rate
declined from 11% in 1981 to 2% in 1999.
26
Starting in the 1990s people
focused their investments predominantly on IT stocks. As stock prices
kept increasing, high rates of return for many short-term oriented venture
capitalists (VCs) attracted more and more investors who were willing to
make a quick fortune themselves. According to Kindleberger and Aliber
(2005), the capital available to VCs as a group surged by a factor of five
during the boom phase. Initial public offerings (IPOs) that usually resulted
from entrepreneurship supported by VCs promised significant capital gains
for everyone involved. In most cases the share prices were significantly
higher after the first day of trading. This demonstration of ever increasing
stock prices, the unstoppable demand for IT stock, overoptimistic experts
and the constant exposure to news and stories about extremely successful
22
See: Schiller, R. J., "Irrational Exuberance" 2nd edition, (2005), p.37 ff..
23
See: Schiller, R. J., "Irrational Exuberance" 2nd edition, (2005), p.39.
24
See: http://www.multpl.com/inflation/table, retrieved 05/15/2013.
25
See: Kindleberger, C. P./ Aliber, R., "Manias, Panics and Crashes" 5th edition, (2005), p.158.
26
See: http://research.stlouisfed.org/fred2/graph/?s[1][id]=PSAVERT, retrieved 05/15/2013.
15
investments made by average people convinced even those who had not
yet believed in the new era talk. Most investors seemed to have ignored
the fact that many of these new IT start-ups had not yet made any profit.
But expectations about possible future gains, or even worse, the prospect
of having missed out on a great money-making opportunity, made most of
them blind to this uncomfortable truth. As already mentioned, when it
comes to short-term investments, market participants extrapolate the
trend forming expectations about prices, projected revenues and the like
based upon recent price movements. In times of a boom or euphoria, this
strategy is extremely dangerous and helps further blowing up the bubble.
The FED (Federal Reserve System) under Alan Greenspan did nothing to
stop or at least dampen the speculation. In 1995 interest rates were
increased last until August 1999. According to Shiller (2005), the market's
interpretation of statements made by the FED was that interest rate
increases were over by March 1995 and that the FED instead prepared for
a soft landing. Kindleberger and Aliber (2005) point out that the market
value of the New York Stock Exchange in 1998 had increased by 40% and
that the NASDAQ Composite (NASDAQ), the market where all the new
technology stocks were traded, had increased by 90% in only twelve
months.
27
They convincingly draw a connection to the imploding Japanese
asset price bubble around that same time. International as well as US
investors restructured their portfolios and started to increasingly invest
into the soaring US economy. The surge of capital inflow into the US
resulted in an increase in domestic investment and a decrease in the US
savings rate, which is equivalent to an increase in investment and
consumption. The US trade deficit increased and the constantly increasing
capital inflow led to an appreciation of the US dollar. Foreign goods
became relatively cheaper for the US, dampening inflation. According to
Kindleberger and Aliber (2005) most of the inflowing funds were used to
either buy more securities or to increase consumption as wealth objectives
were achieved, through the perceived greater wealth made by
27
See: Kindleberger, C. P./ Aliber, R., "Manias, Panics and Crashes" 5th edition, (2005),p.162.
16
investments in the stock market. Investors felt richer and started to
diversify their portfolios turning to housing. The FED began to increase
interest rates starting in August 1999 from below 5% until January 2001
to over 6%.
28
The NASDAQ peaked at 4,572 in January 2000 before
declining and bottoming in June 2002 at 1,172.
29
The Dow Jones
Industrial Average (Dow), where traditional companies like Ford or
General Motors are traded, peaked in January 2000 at 11,723 falling until
March 2003 to 7,674.
30
With the stock market bubble deflating the FED
was fast in aggressively lowering interest rates beginning in January 2001,
putting real interest rates well into the negative
31
, paving the way for the
housing boom. The Dow started its race to new heights in 2003, picking
up speed during 2006 reaching its new high in October 2007, peaking at
14,164.
32
With the stock market folding at high speed, with real interest
rates being negative and with the proclamation of the "ownership society"
under the Bush administration, investors fled the stock market turning to
real estate as an investment option instead. Home prices started to
increase in 1998 and picked up speed in 2000. Prices for homes increased
by 52% from 1997 to 2004
33
, much faster than income increased during
that same period. According to Shiller (2005), people were afraid that
houses soon could be too expensive for them to ever be able to afford
one. Short-term gains and the rush out of the stock market with the belief
that real estate was a safe investment were just two of many possible
psychological causes justifying the new found love for houses. Moreover,
already during the stock market boom in 1990s, there seemed to be
declining standards among managers who were more and more interested
in creative accounting and who appeared to be more concerned with their
short-term benefits rather than long-term profitability of the companies
they worked for. During the housing boom lending and mortgage
standards seemed to be decreasing at a rapid rate. Lending standards in
28
See: http://www.tradingeconomics.com, retrieved 05/17/2013.
29
See: http://finance.yahoo.com, retrieved 05/15/2013.
30
See: http://www.fedprimerate.com, retrieved 05/16/2013.
31
See: Schiller, R. J., "Irrational Exuberance" 2nd edition, (2005), p.41.
32
See: http://www.fedprimerate.com, retrieved 05/16/2013.
33
See: Schiller, R. J., "Irrational Exuberance" 2nd edition, (2005),p.12.
17
the US had changed in a way that formerly not eligible customers were
now able to obtain a mortgage for their homes. The structure of
mortgages had also changed, where traditional mortgages were being
replaced by adjustable-rate mortgages (ARMs), especially after 2003. With
ARMs, initial interest rates are only temporarily low and increase after
some time. As mortgage rates are likely to increase in the future, possible
payment problems, especially when interest rates also increase, are
deferred to a later date. However, under these ARMs, more homeowners
qualified for a mortgage, specifically the lower income, minorities and less
educated people.
34
These mortgages constituted the market for subprime
mortgages. The emergence of a shadow banking system supported by
financial innovations to circumvent regulation is another indicator for
decreasing lending standards. Home prices peaked in 2006 followed by a
drastic decline of 29%-30% through January 2013
35
as many borrowers
were not able to make their mortgage payments. Increased foreclosure
rates in 2006/2007 led to the subprime mortgage crises in the US in 2008
which later morphed into a worldwide financial crisis leaving most
countries in a recession.
34
See: http://www.consumerfed.org, as cited by Schiller, R.J, "Irrational Exuberance" 2nd edition,
(2005),p.211.
35
See: http://www.standardandpoors.com, retrieved 05/16/2013.
18
3.
Common theory and quantifiable facts
"The most challenging difficulty in the study of a financial market is that the nature of the
interactions between the different elements comprising the system is unknown, as is the
way in which external factors affect it."
36
In this section a quick overview of approaches to markets will be given. It
will show how basic theories have changed over time and evaluate the
efficiency of the respective theory in trying to explain extreme events in
these markets.
3.1
Efficient market hypothesis
According to the efficient market hypothesis (EMH), known through
Eugene Fama, all publicly available information is already priced into
financial assets. Any possible up- or downward movements in equity
prices are anticipated. For this reason EMH assumes that financial assets
are always priced correctly which means that the fundamental value and
the value of the respective asset are equal. Only new information can
change the prices which causes a random walk for assets in financial
markets. Prices therefore will always be unpredictable. "The most efficient
market of all is the one in which price changes are completely random and
unpredictable."
37
Correspondingly it should not be possible for any
investor to beat the efficient market. After EMH no investor can buy cheap
and sell dear and make a good profit consistently. Referring to Shiller
(2005) EMH assumes that performance and brilliance do not have any
influence when it comes to investing. This may also be the reason why
many people believe they do not have to pay attention to whether stock
prices are over or underpriced and therefore ignore the usual valuation
of the markets. Sornette (2004) states that the more intelligent and the
harder working investors are, the more random the price changes will be
under EMH. Warren Buffet is the most prominent and probably most
successful example of an investor doing what according to EMH seems
36
See: http://arxiv.org/pdf/cond-mat/9905305v1.pdf, retrieved 05/24/2013.
37
See: Sornette, D., "Why Stock Markets Crash", (2004), p.41.
19
implausible if financial markets were efficient, investing smart, buying
cheap, selling dear and making a good profit. Sornette (2004) found that
in order for stock markets to work properly there have to exist some sort
of noise trades which are less or not informed at all next to the informed
investors. If all investors had the same information then all prices would
be the same as well. If that is the case then obviously there is no
incentive to trade. Further the hypothesis assumes that all investors act
rationally, making any decision by evaluating all available information
correctly. It seems unrealistic for all investors in financial markets to have
the same information on all stocks at all times making decisions only
based on rational calculation considering that the average daily turnover
only on the New York Stock Exchange in 2003 was greater than forty
million US dollars, including more than three million registered trades per
day.
38
Trying to explain economic bubbles seems to be a fruitless attempt
through EMH since upswings in prices caused by speculation should not
exist. Proponents of this theory argue that investors who are not as
informed could cause deviations away from the fundamental value without
reasoning.
39
However, the availability and the processing of the same
information by all market participants is a crucial corner stone of EMH and
should hold. Social interaction and possible feedback loops caused by this
interaction are totally ignored. If markets were frictionless and would exist
in an idealized world with no transaction costs
40
then EMH could possibly
hold. However reality shows that especially stock markets violate the
efficient market hypothesis and that only some fraction in stock market
fluctuations can be justified by new information.
41
Nevertheless EMH has
had a huge influence on policy makers over the past decades while
deregulating markets. Therefore, in analyzing recent economic and
financial evolution in combination with the emergence of respective
bubbles the hypothesis has to be kept in mind.
38
See: http://www.nyse.com, retrieved 05/23/2013.
39
See: http://online.wsj.com, retrieved 05/23/2013.
40
See: Sornette, D., "Why Stock Markets Crash", (2004), p.41.
41
See: Schiller, R.J., "Irrational Exuberance" 2nd edition, (2005),p.193.
20
3.2
Equilibria, bifurcations and chaos theory
Through this very mathematical approach scientists consider economical
questions concerning the existence of (market) equilibria, their
uniqueness and their stability, the adjustment path towards the
equilibrium, its economic properties and alternative system states. The
analysis of differential equations representing a possible market system
nearly always starts out with testing for any equilibrium. Many
multidimensional systems consist not only of one but multiple equilibria,
each having a different influence, positive or negative, on the system. An
equilibrium can be locally or globally stable. If the state variable
converges towards the equilibrium irrespectively of the initial value, then
the system is considered globally stable. If however the state variable
converges only when close to the steady state, it is locally stable. Globally
stable equilibria imply that the system has only one equilibrium. Local
stability implies unique local equilibria. This assumption cannot be met if
there exists a continuum of equilibria. Then, any deviation away from the
equilibrium entails no convergence toward the steady state at any point in
time. Five different stability cases are defined for linear systems. In an
unique globally stable equilibrium, the state variable converges back
towards the steady state, irrespectively of the initial value. A continuum of
instable equilibria implies that none of the equilibria is locally or globally
stable. Any equilibrium can only be reached from its own position. If no
equilibrium exists, then the state variable moves monotonously toward
±. The initial value defines any double cycle. The state variable
alternates between the initial value and some other value respectively.
The steady state is unstable. If a system is uniquely unstable, then the
same accounts for the steady state. The system is defined either by a
monotonous or by an alternating divergence. In summary there are two
possible scenarios in a dynamic system, either convergence or explosion,
with them being monotonous (for linear systems only), alternating or
oscillating. Continuous endogenous movements, either oscillating or
alternating mark exceptional cases. Indicators for the steadiness of a
21
dynamical system are the amplitude and frequency. The amplitude defines
the maximum displacement from the steady state. The frequency is
defined through
, where the period length describes the length
of one cycle, that is its oscillation from maximum to maximum. The larger
the frequency, the shorter the length of a cycle. This observation can be
made in financial markets, where the amplitude and the frequency seem
to have increased over the past decades. This phenomenon also has an
effect on the emergence of economic bubbles which seem to be occurring
in shorter time spans being more severe. An early supporter of the use of
differential equations to describe markets and their movements was
Samuelson with his multiplier-accelerator model (1939). In his linear
model the steady state is consistent with the findings of Keynes, it implies
that an equilibrium always exists. The multiplier and the accelerator do in
fact cause oscillation, however not continuously. In reality this means that
the economy would need constant exogenous shocks for it to develop and
be healthy. The model only considers a closed economy, only works with
persistent endogenous oscillation and may be unstable. Nevertheless it did
motivate modern business cycle theory and is still used today as basic
approach for further evolved models.
Non-linear differential equations defining a system are closer to reality
than linear systems when it comes to describing (financial) markets.
However, they can only be solved in a timely manner with the help of
computer programs. These systems may have no equilibrium, one or
multiple stable equilibria. They may have one or more unstable steady
states or a continuum of equilibria, so called quasi-periodic oscillations.
They may also be defined by chaos. Moreover there may emerge different
combinations like the LAS (locally asymptotic stable) equilibrium or
chaotic movements like the random walk for example. With a globally
stable steady state the system will reach its equilibrium irrespectively
from the initial value. If there exist multiple equilibria, then the steady
states of y
1
and y
3
will be locally asymptotic stable, y
2
will be instable
22
being equal to the initial value. If there does not exist any equilibrium
then the state variable moves toward ±. In order to test a non-linear
system for its locally asymptotic properties, the system has to be
converted into a linear system. The generated linear system can only
approximate the non-linear system well if close, that is local, to the steady
state. The logistic map
, is a
unimodal function and is a great example of how a seemingly simple and
linear equation can, with increasing , turn the system into chaos. Chaos
here means that the timeline will not explode, the system is sensitively
dependent on its initial state, also known as the butterfly effect, and the
movements in the system are irregular but deterministic.
For
there exists one stable steady state at
. For
exist two locally stable (LAS) equilibria. For
there also exist two
equilibria, however both of them unstable resulting in a double-cycle. This
can be seen in the bifurcation above where at the bifurcation point after
the line parts in two, representing a double-cycle. At
one can
observe another partition resulting in a cycle with a length of four, later
Fig. 1.1 Local Bifurcation Diagram of the Logistic
Map, Source:
http://mathworld.wolfram.com
Note for this bifurcation:
23
eight and then, at
chaotic behavior. Bifurcations are a nice way to
illustrate what effect small changes in parameters of formerly stable
systems can have. From one stable steady state, the system can turn into
one where it jumps around between two or more equilibria which are
more or less stable. Similar behavior can be seen for example in the stock
markets. If changes in the parameters are big enough, then the system
may end up in deterministic chaotic behavior. A recent paper in the field
of complex dynamical systems also suggests that there may exist "generic
early-warning signals" just before the state of the system tips when
approaching a critical threshold.
42
Critical slowing down occurs just before
the stable system enters the bifurcation-phase, which may lead to cycles
or chaos later on. A phenomenon seen especially during times of highly
leveraged speculation like in Japan or the US as earlier discussed. A
slowing down in the formerly rapidly increasing prices for real estate and
stock resulted in trend reversal and the tipping of the system. Moreover,
spatial patterns can emerge before a critical transition. It can be observed
that in systems consisting of different units, connected units tend to take
on similar states, increasing the cross-correlation before a critical event.
This phenomenon could be observed in real estate markets around the
world in the early 2000s. Homes seemed to not only be overpriced in the
US but also in the Nordic countries of Europe for example. Housing prices
also surged regionally in Hawaii when prices of real estate were extremely
high in Japan. Moreover, international stock price movements as well as
the recent financial crisis seem to support this theoretical finding. Power-
law structures such as fat tails in returns may vanish as a critical
transition is approached. Recapitulatory, increased trade volatility, or a
volatility calm, systematic relationships between variance and first-order
autocorrelation and increased spatial coherence are, according to Scheffer
et al. (2009) all possible warning signals for approaching critical
42
See: Scheffer, M. et al., "Early-warning signals for critical transitions", (2009).
24
transitions in financial markets. However, the paper stresses that "...there
is no one-size-fits-all spatial pattern announcing critical transitions".
43
Nevertheless, these models, linear or non-linear, are not fully capable of
explaining market movements, especially during times of speculation.
Mostly due to the fact that market psychology does not seem to play too
big of a role here, if it does at all. Data is observed, steady states are
calculated and their stability checked. Although this is an important part of
coming to understand markets, especially the financial ones, it leaves out
one if not the greatest manipulator, the people participating in the
markets itself.
3.3
Agent based financial market models
With increasing computer capacity, this branch of research has evolved
from the non-linear models discussed earlier. Scientists approach financial
markets with the help of stochastic models by imitating complex
dynamical systems. Stylized facts of financial markets like speculative
bubbles and their inevitable burst, high volatility, fat tales in returns
(leptokurtosis), martingales, autocorrelation of returns and volatility
clustering are included in these models to replicate financial market
movements as precisely as possible. One specific interest lies in modeling
the dynamical interactions of different market participants (agents) and
their consequences for financial markets. The financial market models are
designed in so called computer labs and are capable of painting a very
close picture of reality. They can be used to find possible causes for
upswings in the markets or to check which impact for example the
introduction of a tax on financial transactions (Tobin tax) may have,
without actually disturbing the market. The assumption is that movements
in the markets are caused by the interaction of heterogeneous, boundedly
rational agents. Typically, agents are divided into chartists and
fundamentalists. Chartists use technical analysis (TA), fundamentalists
43
See: Scheffer, M. et al., "Early-warning signals for critical transitions",(2009).
25
use fundamental analysis (FA)
44
. Technical analysis assumes that all
relevant information is already included in the market price. The study of
recent price movements is indispensible. It is assumed that prices move in
accordance with trends ("the trend is your friend"). Hence, chartists are
particularly interested in identifying these trends in order to take
advantage of them (trend extrapolation). If prices increase, chartists
believe that markets are undervalued and will continue to go up. They will
therefore start to buy. But if prices decrease, chartists are convinced that
markets are overvalued which is equivalent to a sell signal. Further, TA
assumes that certain structures in the markets are repeated. On that
account chartists will look out for certain formations, like trend lines,
head-shoulder formations, triangles or flags for example to determine
their strategy. Besides these graphical approaches there also exist
quantitative methods that help chartists decide on their market strategy.
These include among others the moving average or the Japanese
candlesticks. However, there seem to exist innumerable rules, which make
chartists' explicit choice appear arbitrary and random. Generally, chartists
cause greater disturbances in the market. Through trend extrapolation
their actions will intensify any upward or downward trend, regardless of
possible fundamentals in the market. However limiting chartists to a
minimum or even crowding them out could trigger an equally great
unrest.
The basic idea of the fundamental analysis is that there exists a
fundamental value over time. The fundamental value follows a random
walk and is typically represented by the Brownian motion (martingale).
Only new information can influence and change the path both, negatively
or positively. Either event is equally likely resulting in the typical ups and
downs in the market which makes its path therefore unpredictable. And
even though the true fundamental value is unknown, or perceived
differently by each agent, there seems to exist a certain trend that prices
44
Market surveys have found that investors equally use TA and FA when deciding on a strategy in
the short-term. For longer horizons FA seems to be preferred. See for example Taylor M.P./ Allan
H. (1992).
Details
- Pages
- Type of Edition
- Erstausgabe
- Year
- 2014
- ISBN (eBook)
- 9783954897025
- ISBN (Softcover)
- 9783954892020
- File size
- 546 KB
- Language
- English
- Publication date
- 2014 (January)
- Keywords
- Possible market policies to avoid extreme movements in the markets Financial and economic bubbles The Japanese asset price bubble Animal spirits – psychological factors for market movements the dot com bubble and the recent financial crisis
