Paul Spallino "The Wealth Effect Hypothesis: a Test in the U.S. Economy"
6
the existence of such relationship in order to forecast future economic consequences that
the wealth effect may lead to.
Until now we have seen a "positive" wealth effect that boosted up the economy because
of the gains achieved, but what would happen if stock markets invert their trends.
According to this hypothesis it should lead to an economic depression. Losses in the
stock market would reduce the wealth of the investors that therefore would have to
moderate their spending. A reduction in consumption will be then be reflected on the
level of output, which could then turn in a fall in share prices causing a further reduction
in investors wealth: in this way the economy would just enter in a vicious circle.
This hypothetical scenario would also be made worse off and amplified by the fact that
the American household have now one of the highest level of private debt in the last 50
years.
This project subject represents a good example of how the real economy and the financial
markets are interconnected together. The accomplishment of this piece of work will
involve the application of a good and complete mixture of microeconomic,
macroeconomic, financial and econometric analysis, evidencing the skills acquired
during the whole course of study.
First a general analysis of the U.S. economy in the 1990's is going to take place in order
to elucidate the background scene in which the wealth effect takes place. The concept of
wealth effect will be introduced. The symptoms, the evidence, and the possible causes
Paul Spallino "The Wealth Effect Hypothesis: a Test in the U.S. Economy"
7
and origins of the phenomenon will be briefly investigated, and will provide the
economic justification of our starting hypothesis.
The third chapter will present an extensive and detailed literature review. The most
relevant theoretical consumption models, such as the Life Cycle Hypothesis and the
Permanent Income Hypothesis, will be introduced, and then a second part will cover the
most recent and specific studies on the "Wealth Effect" in the American situation.
Next an econometric model will be presented and discussed. It will be used in order to
test the validity of the wealth effect hypothesis and quantify its long-run dynamics.
The results will be presented and analysed. A final chapter will conclude.
Paul Spallino "The Wealth Effect Hypothesis: a Test in the U.S. Economy"
8
2 - The U.S. economy in the 1990’s
The way economic policy has been conceived and applied in the past, changed radically
in the last decades.
Since the 2WW, in the previous economic cycles, the demand was stimulated until the
productive capacity was exhausted and inflation rose. A subsequent increase in interest
rates by the monetary authorities would have then followed to disinflate the economy.
The current economic cycle has been different, since the beginning of the 1980's another
conception of economic policy came into the game: the supply-side economics.
The U.S. authorities decided to stimulate the supply more than the demand (the so called
"Reaganomics"). This had been done by taking various actions. In particular the
reduction of the fiscal burden on firms and the introduction of an antitrust legislation that
has facilitated a series of corporate mergers.
In the mean time technology started to make sensible progress, and investments
accelerated remarkably. This situation flowed into a fantastic increase in productivity.
The synergies of key technologies markedly elevated prospective rates of return on high-
tech investments, led to a surge in business capital spending, and significantly increased
the underlying growth rate of productivity. A greater level of productivity increases the
level of non-inflationary rate of growth.
Paul Spallino "The Wealth Effect Hypothesis: a Test in the U.S. Economy"
9
The gross domestic product (GDP) had since 1995 an average annual increase of 4%1.
This was higher than the rest of the world’s economies, and higher than its historical
average of 2.5%. The figure is even more impressing if we look at the whole decade data.
The first quarter of 1990 showed a GDP value of 5720 bln.$, and in the last quarter of
2000 it was 10114 bln.$, representing an increase of 76% in nominal value. Personal
consumption expenditure increased 83% in nominal value and 34% in real value.
Those conditions boosted profits, and low inflation allo wed the Federal Reserve to
maintain an expansionary monetary policy that fuelled investments.
The United States economy in the 1990’s presents remarkable similarities to the
economic situation experienced during the 1920’s (Chancellor, 1999). The rapid
expansion of I.T. technologies stimulated the economic growth, just as the motorcar did
in the 1920’s. Corporate profits were similarly enhanced by the inability of unions to
push through the real wage increases, and workers increased their consumption by
massively using the credit facilities.
This economic environment has also been accompanied by the belief that the economy
was entering into a new paradigm, the so-called “Goldilocks economy”. This theory
suggested that the control of inflation by the Federal Reserve, the decline in federal
deficit, the opening of global markets and the restructuring of corporate America,
1
The sources of all the data presented in this section are:
-Federal reserve web site: www.federalreserve.gov (the section "Data and Statistics")
-the economic data site of the Federal Reserve of St.Luis: www.economagics.com
*It is going to be mentioned when the source of data is different.
Paul Spallino "The Wealth Effect Hypothesis: a Test in the U.S. Economy"
10
accompanied by an exponential innovation in information technology and a massive use
it would have combined to do away with the business cycles.
President Clinton was also influenced by this idea observing that the U.S. economy was
performing so well. The chairman of the Federal Reserve Alan Greenspan and its faith in
a continuous increase in productivity fed the myth of an invulnerable economy and
therefore a strong stock market.
In parallel this development attracted a growing demand of Dollars from foreign
investors creating a positive capital inflow, as well as a strong appreciation of the Dollar.
At the same time the confidence of economic agents was strengthened.
The expectation of higher future profits led to an increasing demand for shares, especially
for the new-economy related sectors (i.e. high-tech and telecommunications). By the end
of 1994 all the main American stock market indexes started showing higher than normal
increases. In order to give an idea of the magnitude of this increase we can mention the
fact that the Dow Jones Industrial Average, the index that groups the 30 companies with
the highest stock market capitalisation (Large caps in financial jargon), rose from a low
of 2,3652 index points in 1990 to 11,800 in 2000, representing a gain of 400% in nominal
value, and 270% in real value (deflated with the CPI all goods index).
The Nasdaq composite index has been as high as 5100 in the beginning of 2000. Its value
in the first quarter of 1990 was 435.
By the way these dramatic increases in stock market figures do not seem to reflect the
"real economy" performance if compared with GDP levels. Moreover if we look at
2
The source of all data regarding stock markets is: Bloomberg software
Paul Spallino "The Wealth Effect Hypothesis: a Test in the U.S. Economy"
11
corporate profits figures, which showed an increase of 140% during the 1990's, this
overvaluation seems to be evident.
Most analysts who have examined this stock market prices in light of perspective future
profit growth have concluded that the stock market was vastly over valued (Baker 1997,
Diamond 1999, Schiller 2000). Especially the period between 1995 and 1999 has
experienced a bubble not sustainable in the long run. This argument was supported by the
fact that the ratio of stock prices to corporate earnings (P/E ratio) peaked in 1999 to more
than thirty to one. This ratio is more than twice its historic average, which has been
approximately 14.5 to one over the last fifty years.
In parallel the level of personal debt has been increasing, and most of the money
investors have been investing in the stock market is money they have borrowed.
American households have now one of the highest level of private debt in the last 50
years. The figures about total consumer credit outstanding doubled their value during the
decade getting to 1534 bln. $ at the end of the 2000.
Increasing debt caused a strong growth and investment in shares (Chancellor, 1999).
As in the1920’s, the1990’s stock market bubble has been stimulated in particular by the
low interest rates policy applied by the Federal Reserve since the beginning of the
decade.
In addition to that there was a widespread public belief that private investment in shares
would have supported the stock market in the long term. This faith in the stock market
was accompanied by the belief that investments in shares would provide higher returns
compared to investments in obligations. This idea led investors to buy shares without any
Paul Spallino "The Wealth Effect Hypothesis: a Test in the U.S. Economy"
12
consideration for the level of prices. In 1998 the average P/E ratio (price/earning) rose to
its historical peak of 30, as mentioned before.
The only financial risk considered in the 1990’s was to leave money in saving accounts,
while the stocks were rising at a 20% annum. In addition to that, most of the share
purchases were financed by leverage d systems. Between 1990 and 1998 credit margins
increased five times.
It is important to note that in this period the popular participation to the markets increased
sensibly, from 1994 to 1997 the aggregate value of household sector equity holdings
roughly doubled. The economic and psychological background scenario previously
presented justify in a certain way the fact that households started introducing investment
practices into their private wealth management, thus becoming an active part in the
dynamics of the financial markets.
The explosion of investment trusts in the 1920’s has been neatly exceeded by the rapid
growth of mutual funds in the 1990’s. Between 1990 and 1998 equity funds attracted
over a trillion dollars, and their number changed from 1100 to over 6,000.
In 1996 a total of $221.6 billion was invested in mutual funds, and the same amount in
1997. By the beginning of 1998 the total assets of U.S. mutual funds had risen to $4.2
trillion, a sum roughly equal to the assets of the banking system.
In the middle of the 1990’s, the existence of an “equity cult” became evident (Chancellor,
1999). This social phenomenon has been favoured in its practical aspect also by the
development of I.T. tools, in particular the internet and trading online services.
Paul Spallino "The Wealth Effect Hypothesis: a Test in the U.S. Economy"
13
More than 50 millions American households held shares and the stock market became a
very spread topic of conversation. By 1998 there were over 37,000 investment clubs,
while those were only 6,000 at the beginning of the decade.
However, what we have discussed so far, only present us the scenario of a boosted
economy with strong fundamentals, but still overvalued. The fact that private household
investments skyrocketed, suggests that the level of private wealth became more
vulnerable and likely to be influenced by the trends shown in the stock markets.
The element that should attract now our attention is now the fact that during this exact
period, the 1990’s, macro-aggregate consumption patterns started to show a progressive
increase, not justified by increases in the level of personal disposable income.
If we have a look at the consumption to disposable income ratio (C/Y), that tell us the
proportion of income spent in consumption, we can see that at the beginning of the
decade it showed a value of 0.89, while at the end of the year 2000 its value was 0.97, an
increase of 9% (author’s calculations). This data shows a very strong correlation with the
stock markets trend. Using the Whilshire 5000 Total Market Index, the American index
that groups all the securities quoted in the NYSE (New York Stock Exchange), as
representing stock markets trends and correlating it with the consumption to disposable
income ratio the result is 0.90 (author’s calculation). This means that movements in the
Whilshire index explain 90% of movements in C/Y ratio. Figure 1 graphically represents
the parallelism between the two trends, by outlining the annual changes of both.
Paul Spallino "The Wealth Effect Hypothesis: a Test in the U.S. Economy"
14
Fig.1 Data source: C and Y from www.economagics.com; Whilshire 5000 from www.whilshire.com
This data does not mean in itself that there is any causal relationship between the two sets
of data. Instead, it gives us evidence that the two movements have been parallel and
further investigation should be actuated. This relationship represents the so-called
“Wealth Effect”.
If this effect were true and had a strong magnitude, the traditional dynamics that governs
the relationship between “real world economy” and the stock market would change.
Normally, a strong economy with an high rate of productivity, technologic innovation
etc., is represented by strong stock market valuations. Vice-versa when there is a
downturn the market goes down.
Let us imagine a strong economy that has strong stock markets, in which everyone is
optimist and every one invest in it. Optimism and positive expectations accompanied by a
Whilshire 5000 vs. C/Y ratio
(annual % change)
-30
-20
-10
0
10
20
30
40
19
90
19
90
19
91
19
92
19
93
19
93
19
94
19
95
19
96
19
96
19
97
19
98
19
99
19
99
20
00
Years
%
Ch
an
ge
Whilshire %Ch. C/Y ratio % Ch.
Paul Spallino "The Wealth Effect Hypothesis: a Test in the U.S. Economy"
15
big popular participation cause higher securities valuations, which in practical terms are
translated into capital gains. Those capital gains increase household's wealth. People feel
richer because of the increased value of financial assets and are likely to change their
consumption behaviour by increasing their spending. By increasing their spending,
consuming more in aggregate terms, households boost the real economy. Aggregate
demand increases and companies sell more. Higher profits are generated and companies'
quotations are likely to increase further.
A virtuous circle is then created, such that rising prices in the stock market are justified
with beliefs and theories that cause a further increase. As John Maynard Keynes argued
back in the 1930’s, a booming economy is based in good part on the expectation that the
economy will continue to boom.
In the short term this is useful for the economy. Consumers spend their financial gains to
buy more shares, and ignore their debts, firms issue new shares and bonds to purchase
other companies or finance capital expenditure, and the government enjoys rising tax
receipts as the economy prospers. The stock market is seen as the greatest creator of
wealth, a “perpetual motion machine” (Chancellor, 1999).
The first problem rises form the fact that the money spent in such situations is "virtual
money". People are virtually richer because of the capital gains, but until the shares are
not sold these gains are not liquid. If all the investors were to sell their securities the
prices would suddenly depreciate. So where does this money come from?
Paul Spallino "The Wealth Effect Hypothesis: a Test in the U.S. Economy"
16
In primis a greater part of disposable income is spent. Saving decreases since it is counter
balanced by the higher level of wealth. The average value of savings for year 2000 in the
U.S. has been negative (-8.4 bln. $).
The second source is credit. The higher value of financial assets makes it easier to
borrow money by increasing the value of principals.
The second real problem is that unfortunately at a certain point the limits of an economy
are reached. Business cycles reappears and the perpetual motion machine turns direction.
A loss of faith in economic agents positive expectation can set off a vicious cycle.
Concerns about the fact that this might happen in a near future are justified by the latest
evidence.
If we look at all the main stock market indexes the situation is a bit controversial. The
Nasdaq index burst its bubble and has been in crisis since the second half may falling as
low as 1600 in the beginning of April 2001 from its peak of 5100 in the first half of the
2000, a drop of 219%. The Dow Jones index lost some points, but until now (April 2001)
it has been moving around an average value of 10,000 points. The bubble is still
substantially intact considering that at the beginning of the 1990's it was at 2000, and that
since then the GDP increased of 140%. Moreover if we look at the average P/E ratio of
quoted companies it still maintain itself at a level of 20-25 to 1, which still is higher than
the historical average. This is actually surprising considering the fact that during the
course of this year all the main companies (among which: Intel, Cisco Systems, IBM,
General Motors, etc.) have given warnings announcing a slowdown in sales and earning
growth.
Paul Spallino "The Wealth Effect Hypothesis: a Test in the U.S. Economy"
17
The slowdown in production has been already visible in the latest GDP figures. From the
last quarter of 1999 to the last quarter of 2000, annual gross domestic product growth
tumbled from a supercharged 8.3% to just 1.1%, the most rapid decline since WW2.
Moreover certain economists expected to see economic growth turning into a negative
value in the first quarter of 2001 (Coy, 2001). The actual data for GDP growth in the first
quarter of 2001 (announced in April 27, 2001) showed a positive value of 2%. This data
has been higher than the expectations, but still shows a significant slowdown.
The fact that GDP is still positive might be justified by the fact that personal consumption
expenditure is holding. The figure for February 2001 was $7003.5 bln. Which represents
an increase of 1.6% since December 2000. At first sight consumption does not seem to
have suffered immediately from the stock market crisis. This does not exclude the fact
that there might be a lagged response.
On the other hand the Federal Reserve is maintaining a low interest rates monetary policy
to maintain consumption levels positive. The interest rate has been recently (April 2001)
lowered at 4%, and it is now 200 basic points lower than last year.
This fact shows how the monetary authorities are worried about a potential fall in
consumption. Such an event would hit production levels, worsen the whole situation
causing a "hard crush" of the American economy. This would be caused partly by the
"Wealth Effect".
So far we have then seen what is happening in the U.S. in this period, and we have
detected the "Wealth Effect" as a potential threat for its economy.
Paul Spallino "The Wealth Effect Hypothesis: a Test in the U.S. Economy"
18
Next section is going to present some of the most important studies on the topic, so that
we can see the methodology applied, and the results that these offered.
Paul Spallino "The Wealth Effect Hypothesis: a Test in the U.S. Economy"
19
3 - Literature review
3.1 - Traditional consumption models
To frame the discussion about the "Wealth Effect Hypothesis" it is useful to review the
main models that study, investigate and analyse how an unexpected increase in wealth
would affect consumption. Even if the stock market wealth factor is not explicitly
incorporated into these models, they still leave some space for its consideration.
The relationship between consumer's expenditure and disposable income is one of the
most researched topics in quantitative economics, the following theories are very
representative and have inspired most of the following literature.
The idea of an aggregate consumption function dates from the writings of John Maynard
Keynes in the 1930’s. Keynes posited a fundamental rule: while consumption demand is
determined by many things, it is first and foremost a function of income, consumption
increases with income but consumption as a percent of income is a decreasing function of
income.
Ricevi informazioni sui nostri servizi, sulle offerte e non perdere news e consigli su università e lavoro.
