Review of Economics & Finance
Submitted on 03/11/2014
Article ID: 1923-7529-2015-02-01-18
Andrew R. Blair, Gershon Mandelker, Thomas L. Saaty, and Rozann Whitaker
Forecasting the Resurgence of the U.S. Economy in 2010:
An Expert Judgment Approach *
Andrew R. Blaira),
Gershon Mandelkerb),
Thomas L. Saatyc),
Rozann Whitakerd)
(blair@katz.pitt.edu) (mandelke@katz.pitt.edu) (saaty@katz.pitt.edu) (rozann@creativedecisions.net)
a),b),c): Katz Graduate School of Business, University of Pittsburgh, Pittsburgh PA 15260, U.S.A.
d): Creative Decisions Foundation, 4922 Ellsworth Avenue, Pittsburgh PA 15213, USA
Abstract: This paper describes a forecast of the date of the recovery of the US economy from the
contraction that began in December 2007. The forecast used an expert judgment approach
(featuring 29 pairwise comparisons of the key elements in our forecasting model) within the
framework of decision theory, the Analytic Hierarchy Process, as well as its generalization to
dependence and feedback, the Analytic Network Process. This paper has once again demonstrated
how the Analytic Hierarchy Process can serve as an additional tool for providing macroeconomic
forecasts. We have based our forecast of the time period of the trough of the current economic
cycle within the context of the macroeconomic conditions confronting the U.S. economy during late
2008, which had begun to experience a recession from its peak in December 2007, after an
expansion of 73 months. We have concluded that the recovery would begin during the third quarter
of 2010, from a trough that would be reached some 20 months forward from the date of the
December 2008 forecasting exercise (and some 32 months from the previous cyclical peak, which is
considerably longer than all other post-World War II contractions).
Keywords: Macroeconomic forecasting; Analytic Hierarchy Process; Expert judgment approaches
to economic forecasting; Forecasting U.S. economic expansions
JEL Classifications: C53, E32, E37
1. Introduction
Building on work done earlier, Blair, et al. (2001) and Blair, et al. (2002), this paper illustrates
our use of the Analytic Hierarchy Process (AHP, Saaty (2005) and Saaty (2006) to produce a
December 2008 forecast of when the U.S. economy would begin to recover from the contraction
that, according to an announcement dated December 1st, 2008, from the Business Cycle Dating
Committee of the National Bureau of Economic Research (NBER) (2008), began during the month
of December, 2007. Our earlier forecast of the economic recovery, produced in the year 2001, Blair,
*The original version of this article was initially written during 2009, on the basis of a forecasting
exercise conducted in December 2008, and was published in Socio-Economic Planning Sciences (44
(2010), pp. 114-121), Elsevier Ltd. The current paper is a summary of that earlier article, including an
updated postscript setting forth what transpired after we concluded our forecasting exercise.
~1~
ISSNs: 1923-7529; 1923-8401 © 2015 Academic Research Centre of Canada
et al. (2002), was confirmed by the NBER in July 2003 as reported in the Wall Street Journal by
Hilsenrath (2003). As previously, we employed a conceptual framework grounded in modern
macroeconomics and produced a forecast that relied upon expert judgment, without assistance from
a conventional macroeconomic forecasting model.
2. The Role of Judgment in Economic Forecasting
As stated in our earlier articles, conventional approaches to macroeconomic forecasting tend to
be constrained by the estimated values of parameters and intercept terms. These are imbedded in
the multi-equation models that are typically employed to produce "first-cut" forecasts of relevant
endogenous variables. Additionally, the values of a large number of "exogenous" variables
(relating to the future course of monetary and fiscal policy, the value of exports, etc.) must be
subjectively estimated on the basis of available evidence and consensus judgment. Initial forecasts
produced by the raw models are then typically adjusted by "add" or "fudge" factors, most
commonly in the form of shifts in the values of previously estimated intercept terms. This
procedure is employed in order to produce forecasts that are consistent with recent values of key
endogenous variables when it is evident that a shift of some kind has occurred in portions of the
underlying model structure. Such exercises also provide ample opportunity for resetting the values
of exogenous variables.
Thus, and as stated in our earlier papers, this suggests that macroeconomic model builders/
forecasters are well aware of the limitations of their underlying models and the need to incorporate
subjective judgments. However, these judgmental adjustments are necessarily non-systematic and
ad hoc in nature. Accordingly, here we again utilize an alternative, systematic approach – AHP − in
order to remedy this deficiency. And while we have not illustrated this alternative by adapting a
formal macroeconomic forecasting model, the conceptual framework, as noted above, is grounded
in modern macroeconomics. Our alternative approach, moreover, could also be readily employed to
enrich forecasting exercises based on formal models (e.g. generating add factors more
systematically and consistently; adjusting the values of exogenous variables). As stated in the
earlier papers in this respect the two forecasting approaches can be seen to converge quite
compatibly.
3. The Setting: An Economic Contraction
After Six Years of Expansion
While in popular accounts it is conventional to view the U.S. economy as being in a recession
if real Gross Domestic Product (GDP) has declined for two consecutive quarters, the National
Bureau of Economic Research, utilizing a panel of experts (i.e., the above noted Business Cycle
Dating Committee), has, by general consensus, been given the responsibility for dating the actual
turning points in the U.S. economic cycle. This organization arrives at its assessments by utilizing a
number of indicators of economic activity. It describes a recession as “a significant decline in
economic activity spread across the economy, lasting more than a few months, normally visible in
production, employment, real income, and other indicators,” National Bureau of Economic
Research, Business Cycle Dating Committee (2008). In its announcement, it notes that the
committee views the payroll employment measure, which is based on a large survey of employers,
as “the most reliable comprehensive estimate of employment” and observes that this series reached
its peak in December 2007 and has declined every month since. It views the quarterly estimates of
~2~
Review of Economics & Finance, Volume 5, Issue 2
real Gross Domestic Product and real Gross Domestic Income as normally “the two most reliable
comprehensive estimates of aggregate domestic production.” However, while the two measures are
conceptually identical, measurement issues have produced a statistical discrepancy that has caused
them to behave differently in recent quarters, with ambiguous implications for the dating of the
peak of the most recent economic cycle. However, “Other series considered by the committee –
including real personal income less transfer payments, real manufacturing and wholesale and retail
sales, industrial production, and employment estimates based on the household survey – all reached
peaks between November 2007 and June 2008.” This led the committee to conclude that “All
evidence other than the ambiguous movements of the quarterly product-side measure of domestic
production” confirmed that the contraction had begun at the end of 2007. Moreover, the NBER
dates the trough of the previous cycle as having occurred in November of 2001, with the subsequent
expansion hitting its peak, as we have now been informed, in December of 2007. This 73-month
expansion was considerably shorter than the previous 120-month expansion that occurred between
March 1991 and March 2001, the longest expansion in the post-world war II era, and was also
shorter than two other post-war expansions (February 1961-December 1969; November 1982-July
1990).
The most recent contraction has also been accompanied during 2008 by increasing turbulence
in U.S. and global financial and equity markets (with its onset dating from the middle of 2007),
reflecting, in turn, the fallout from the collapse of the housing market, the sub-prime mortgage
market and related financial instruments such as mortgage-backed securities and credit default
swaps. Additionally world oil prices rose very rapidly throughout much of 2008. For example,
while the widely-quoted UK Brent Blend spot price of oil had been trending upward since the
middle of 2004, it jumped to over $90 a barrel at the end of December 2007 and topped $140 a
barrel in early July 2008. Since then, oil prices have fallen precipitously – to below 40 in December
2008 – as the slowing of the global economy and the attendant diminished demand for oil have
exerted their influence. For much of the year, the U.S. dollar was also under attack vis-à-vis the
Euro and other major currencies, in a pattern that seemed, to some observers, to reflect a run-up in
oil futures prices and the possibility that such futures were being used as a currency hedging
vehicle, in addition to tracking underlying supply and demand factors in the oil market.
The recent economic news has been dominated by the U.S. Treasury’s attempt to prop up the
financial markets with an almost $350 billion infusion of support that was authorized by Congress
as part of a total financial “bailout” package of $700 billion (the remainder of which is being
reserved for the use of the incoming Obama administration). Various measures implemented by the
Federal Reserve System resulted in the pumping of billions of dollars of reserves into the banking
system and a steep reduction in the federal funds rate. Two prominent investment banks were
permitted to be acquired by stronger financial firms (while the bankruptcy of still another prominent
investment bank was allowed to take place without a rescue effort), and two other prominent
investment banks were converted into bank holding companies that then came under Federal
Reserve jurisdiction. Additionally, plans were being laid by the incoming administration for a
major fiscal stimulus package to be implemented soon after taking office (with estimates being
circulated on the order of $500 billion to $1 trillion). Finally, in December 2008, Congress failed to
adopt a proposed package of loans to the three U.S. automobile companies, two of whom were
confronting near-term serious cash flow problems. On December 19 the administration announced
that $13.4 billion from the still unused “first tranche” of the financial bailout package would be
deployed by the administration as a “bridge loan” to the two cash-strapped producers, so that
fundamental issues surrounding the domestic industry could be addressed more directly by the
incoming administration.
~3~
ISSNs: 1923-7529; 1923-8401 © 2015 Academic Research Centre of Canada
4. Application of AHP to the
Macroeconomic Forecasting Problem
Our forecasting exercise employed the AHP to address the timing of the expected resurgence
by seeking to answer the question “what is the most likely period in the future when the resurgence
will occur?” By this term, we implicitly meant a recovery from the trough that will eventually be
confirmed by the NBER, using the kinds of broad measures of the economy noted above. Like
typical forecasters, we were not precise with regard to the projected rate of the expansion.
4.1 Decomposition of the Problem Hierarchically
Decomposing the problem hierarchically, the top level of the exercise consists of the primary
factors believed by us to represent the forces or major influences driving the economy in the fall of
2008: “Aggregate Demand;” “Aggregate Supply;” and “the Current Global Financial Context.”
Each of these primary categories was then decomposed into sub-factors represented in the second
level. Under Aggregate Demand, we identified consumer spending, net exports (i.e., exports less
imports), business capital investment, shifts in consumer and business investment confidence, fiscal
policy, monetary policy, and expectations with regard to the future course of inflation (which, in
turn, could reflect expectations concerning the future course of monetary policy and fiscal policy).
As in the previous exercise, we make a distinction between shifts in consumer and business
investment confidence and the formation of expectations regarding future economic developments.
Under Aggregate Supply, we identified labor costs (which, in turn, are driven by changes in
such underlying factors as labor productivity and real wages), natural resource costs (e.g., the price
of oil and other energy costs), and expectations regarding such costs in the future.
With regard to the Current Global Financial Context, we identified as the principal subfactors
the likelihood of changes in major international political relationships (such as wars or other
international conflicts), the high degree of global financial integration, a number of issues
surrounding the current mortgage markets crisis, expectations with regard to future oil prices, and
the future value of the dollar. We decomposed the current mortgage markets crisis into five subfactors: uncertainty about housing prices, uncertainty about the value of mortgage-backed securities,
the role of credit default swaps, the role of government intervention in the financial markets, and the
lack of confidence in the accuracy of financial reporting. With regard to the various sub-factors, we
recognized that they are, in some instances, interdependent. This is especially evident with regard to
the sub-factors listed under the mortgage markets crisis.
The lowest level of the hierarchy consists of the alternate time periods in which the resurgence
might occur as of December 5, 2008 (preliminary exercises had been conducted during October and
November, 2008): within six months, within twelve months, within twenty-four months, or within
thirty-six months (or beyond). Because the primary factors and associated sub-factors are timedependent, their relative importance had to be established in terms of each of the four alternative
time periods. Thus, instead of establishing a single goal as one does for a conventional hierarchy,
we used the bottom level time periods to compare the Primary factors at the top. This entailed the
creation of a feedback hierarchy known as a "holarchy" in which the priorities of the elements at the
top level are determined in terms of the elements at the bottom level, thus creating an interactive
loop. Figure 1 provides a schematic representation of the hierarchy we used to forecast the timing
of the economic resurgence. There are five clusters, Primary Factors, Aggregate Demand Factors,
Aggregate Supply Factors, Clobal Financial Context and Alternative Time Periods.
~4~
Review of Economics & Finance, Volume 5, Issue 2
Primary Factors
1 Aggregate Demand
2 Aggregate Supply
3 Global Financial Context
1 Aggregate Demand Factors
3 Global Financial Context
1 Consumption
2 Net Exports
3 Investment
4 Confidence
5 Fiscal Policy
1 Major Int抣 P oliticalR elationships
2 Global Financial Integration
3 Mortgage Crisis Issues
Uncertainty about Housing Prices
Uncertainty about Mortgage
Backed Securities
Role of Credit-Default Swaps
Gov抰 O w nership and Intervention
Lack of Confidence in Financial
Reporting
Tax Policy
Gov抰 E xpenditure
6 Monetary Policy
7 Expected Inflation
4 Expectations of Future Oil
Prices
5 Future Value of the Dollar
2 Aggregate Supply Factors
1 Labor Costs
2 Natural Resource Costs
3 Expectations
Alternative Time Periods
1 Six months
2 Twelve months
3 Twenty four months
4 Thirty six months
Figure 1. Overall View of the Structure of the Forecasting Model
4.2 Pairwise Comparison
After decomposing the problem hierarchically, the next step in the process was to pairwisecompare the relative importance of the primary factors (Aggregate Demand, Aggregate Supply, and
the Global Financial Context) as they influence (1) the timing of the economic resurgence; (2) the
relative importance of each of the sub-factors as drivers of the associated primary factor in the next
level of the hierarchy; and (3) the relative importance of each of the subfactors under each primary
factor as it influences the timing of the economic resurgence. These comparisons were carried out
using the nine point fundamental scale of the AHP.
~5~
ISSNs: 1923-7529; 1923-8401 © 2015 Academic Research Centre of Canada
The judgments with regard to identification of factors and sub-factors, as well as the
comparisons of their relative impact and strength, were conducted by the authors, who assumed the
role of representative "experts". Obviously, the outcomes are strongly dependent on the quality of
those judgments. As noted, the final version of the exercise was conducted on December 5, 2008.
Twenty-nine sets of judgment matrices were generated in this exercise. The priorities that
emerged from the exercise for the major factors, sub-factors and their relative importance with
regard to the time frames for a possible turnaround, are shown in the tables in the Appendix. At the
beginning of the Appendix a brief explanation is given about how the questions were posed to elicit
the judgments for each set of pairwise comparisons. A perusal of the tables reveals the following
about the priorities derived from the pairwise comparison judgments:
With regard to attaining an economic resurgence within the six- and twelve-month forecasting
periods, Aggregate Demand factors were judged to dominate the Aggregate Supply and Global
Financial Context factors (79% in each period); for the longer 24- and 36-month time horizons,
Aggregate Demand and Aggregate Supply factors were judged to be of equal weight (45% in each
period), with likely adjustments in the Global Financial Context assuming somewhat greater
relative importance over these longer periods.
With regard to promoting economic resurgence, the consumption, monetary policy, fiscal
policy, and confidence sub-factors were assigned relative weights totaling 89% of the Aggregate
Demand primary factor.
Of the Aggregate Demand sub-factors, consumption, net exports, investment, and confidence
were judged to be most influential in the 24- and 36-month forecasting horizons, whereas monetary
and fiscal (especially government expenditure) policy exerted greater influence in a 12- or 24month time frame.
Of the Aggregate Supply sub-factors, natural resource costs were dominant at almost 60%, and
this sub-factor assumed greater relative importance within the 24- and 36-month time horizons.
Of the Global Financial Context sub-factors, the mortgage crisis issues and the existence of
global financial integration were dominant, totaling 69% of relative importance, with mortgage
crisis issues alone accounting for 42%. The underlying mortgage issues sub-factors themselves
were judged to be generally more amenable to solution within a 24-month or 36-month time
horizon.
Each judgment matrix has an associated priority vector or vector of weights. (These are the
numbers that appear in the supermatrix, Table 3-a,b,c, in the Appendix.) The limit supermatrix
(Table 4-a,b,c in the Appendix) is the result of raising the supermatrix to large powers to
approximate to the limit to which it converges. In this case, the powers of the supermatrix perform a
cycle, and for the overall limit, the sum of the various limiting cycle phases is taken to obtain the
outcome. This is the final supermatrix of the results. The resulting final priorities for the
alternative time periods are obtained from the last four rows of any column in Table 4-a,b,c by
normalizing the four numbers: 0.0307, 0.0634, 0.0953, and 0.1281. The resulting final priorities for
the time periods are shown in Table 1 below: six months 0.0968; twelve months 0.1997; twentyfour months 0.3001; and 36 months 0.4034.
~6~
Review of Economics & Finance, Volume 5, Issue 2
5. Producing the Forecast of the Recovery
To obtain our forecast, we subsequently multiplied each priority by the midpoint of its
corresponding time interval and added the results (as one does when evaluating expected values) as
shown in Table 1.
Table 1. Expected value computation of number of months to turnaround
Midpoint of time period
Time Period Alternative (and
Priorities measured in months; the Priority ×
period covered)
(Normals) starting time when the
Midpoint
prediction was made is 0
(months)
1 Six months (0 to 6)
0.0968
3
0.2905
2 Twelve months (6 to 12)
0.1997
9
1.7970
3 Twenty four months (12 to 24) 0.3001
18
5.4014
4 Thirty six months (24 to 36)
0.4034
30
12.1009
Sum
19.5898
We interpreted this to mean that the recovery would occur about 20 months from the time of
the final forecasting exercise on December 5, 2008, or around early August 2010; that is to say,
toward the middle of the third quarter of 2010. Interestingly, as this paper was drafted in December
2008, a number of private and official forecasters were making projections that were more
optimistic, believing that the trough of the cycle would be reached about a year earlier, Federal
Reserve Bank of Philadelphia (2008) and Organization for Economic Cooperation and
Development (2008).
6. Conclusion
This paper has once again demonstrated how the Analytic Hierarchy Process can serve as an
additional tool for providing macroeconomic forecasts. We have based our forecast of the time
period of the trough of the current economic cycle within the context of the macroeconomic
conditions confronting the U.S. economy during late 2008, which had begun to experience a
recession from its peak in December 2007, after an expansion of 73 months. We have concluded
that the recovery would begin during the third quarter of 2010, from a trough that would be reached
some 20 months forward from the date of the December 2008 forecasting exercise (and some 32
months from the previous cyclical peak, which is considerably longer than all other post-World War
II contractions). Other private and official forecasters are also forecasting a longer contraction than
has been the case in previous postwar cycles, but our forecast is even less optimistic, and reflects
our view that fundamental dislocations have occurred that will take additional time to overcome.
~7~
ISSNs: 1923-7529; 1923-8401 © 2015 Academic Research Centre of Canada
Postscript
On September 20, 2010, The Business Cycle Dating Committee of the National Bureau of
Economic Research announced (2010) that it had “determined that a trough in business activity
occurred in the U.S. economy in June 2009. The trough marks the end of the recession that began in
December 2007 and the beginning of the expansion. The recession lasted 18 months, which makes
it the longest of any recession since World War II.” Thus, this widely-acknowledged arbiter of the
dates for the turns in the business cycle dated the beginning of the current recovery some five
quarters earlier than our own (third quarter of 2010) forecast. While acknowledging that the trough
in “labor market indicators” (aggregate hours and employment) occurred “later” (e.g., “the NBER
trough date is 6 months before the trough in payroll employment”), it based its conclusion on “the
“strong growth of quarterly real GDP and real GDI in the fourth quarter” of 2009, along with the
committee’s belief “that these quarterly measures of the real volume of output across the entire
economy are the most reliable measures of economic activity.”
Further, “in previous business cycles, aggregate hours and employment have frequently
reached their troughs later than the NBER’s trough date.” Thus the trough in labor market
indicators occurred at a point in time that was closer to our own forecast of the turn in the overall
cycle. In any event, using either date, the strength of the recovery has been painfully slow along a
variety of measures (Center on Budget and Policy Priorities (March 2015)), suggesting that
structural elements are still at work that have contributed to a continued slackness in the overall
labor market (despite a growth in employment and a reduction in the formal unemployment rate)
and a failure of real wages to grow significantly during the current expansion.
~8~
Review of Economics & Finance, Volume 5, Issue 2
References
[1] Blair, A.R., Nachtmann, R. and Saaty, T. L. (2001). “Incorporating Expert Judgment in
Economic Forecasts: The Case of the U.S. Economy in 1992,” Chapter 12 in the book: Saaty,
T.L. and L.G. Vargas, Models, Methods, Concepts and Applications of the Analytic Hierarchy
Process, London: Kluwer Academic Publishers.
[2] Blair, A.R., Nachtmann, R., Saaty, T.L. and Whitaker, R. (2002). “Forecasting the Resurgence
of the US Economy in 2001: An Expert Judgment Approach,” Socio-Economic Planning
Sciences, 36(2): 77-91; and also Chapter 2 in Saaty, T.L. and L.G. Vargas, Decision Making
with the Analytic Network Process: Economic, Political, Social and Technological Applications
with Benefits, Opportunities, Costs and Risks, Springer, New York.
[3] Blair, A.R., Mandelker, G., Saaty, T.L, and Whitaker, R. (2010). “Forecasting the Resurgence of
the U.S. Economy in 2010: An Expert Judgment Approach,” Socio-Economic Planning
Sciences, 44(3): 114-121.
[4] Blanchard, O. (2006). Macroeconomics (fourth edition), Prentice Hall, Englewood Cliffs, NJ.
[5]Center on Budget and Policy Priorities (2015). “Chart Book: The Legacy of the Great
Recession,” updated March 9, 2015. [Online] Available at http://www.cbpp.org .
[6] Fair, R.C. (1984). Specification, Estimation, and Analysis of Macroeconometric Models,
Harvard University Press, Cambridge, MA.
[7] Federal Reserve Bank of Philadelphia (2008). “Fourth Quarter 2008 Survey of Professional
Forecasts,” (November 17, 2008), [Online] Available at www.philadelphiafed.org .
[8] Hilsenrath, Jon E. (2003). “Despite Job Losses, the Recession is Finally Declared Officially
Over,” The Wall Street Journal (July 18, 2003).
[9] National Bureau of Economic Research, Business Cycle Dating Committee (Dec. 1, 2008).
“Determination of the December 2007 Peak in Economic Activity,” (Dec. 1, 2008), [Online]
Available at NBER web site http://www.nber.org .
[10] National Bureau of Economic Research Business Cycle Dating Committee (September 20,
2010). Announcement of the June 2009 trough of the current business cycle. [Online]
Available at NBER web site http://www.nber.org .
[11] National Bureau of Economic Research (2012). “Business Cycle Expansions and
Contractions,” [Online] Available at http://www.nber.org/cycles/US_Business_Cycle_
Expansions_and_Contractions_20120423.pdf .
[12] Organization for Economic Cooperation and Development (2008). “Economic Outlook No.
84,” (November 2008) [Online] Available at OECD web site www.oecd.org .
[13] Saaty, T. L. (2005). Theory and Applications of the Analytic Network Process: Decision
Making with Benefits, Opportunities, Costs and Risks, RWS Publications, Pittsburgh, PA.
[14] Saaty, T. L. (2006). Fundamentals of Decision Making and Priority Theory with the Analytic
Hierarchy Process, RWS Publications, Pittsburgh, PA.
~9~
ISSNs: 1923-7529; 1923-8401 © 2015 Academic Research Centre of Canada
Appendix:
The Supermatrix and Its Limit and the Judgment Matrices
Twenty-nine pairwise comparison judgments were performed and the priorities derived from
them were used to form the supermatrix shown in three parts in appendix Tables 3-a to 3-c. The
stochastic supermatrix is raised to powers until it converges; that is, its values remain the same from
one power to the next resulting in the limit supermatrix shown in appendix Tables 4-a to 4-c.
Posing the question when comparing criteria
In this paper we use the terms criteria and factors interchangeably. The caption for appendix
Table 5, for example, reads “The Judgments for the Subfactors of Aggregate Demand with respect
to the Aggregate Demand main factor”. The Aggregate Demand main factor is the parent element
with respect to which its six subfactors in the Aggregate Demand cluster in Figure 1 are compared.
They are Consumption, Net Exports, Investment, Confidence, Fiscal Policy, Monetary Policy and
Expected Inflation. The Fundamental Scale of the AHP is used to express the pairwise comparison
judgments: [1-Equal, 3-Moderate, 5-Strong, 7-Very Strong, 9-Extreme; numbers in-between and
decimals are also allowed]. The question posed is: “Which subfactor is more important in
determining the time to recovery with respect to the Aggregate Demand main factor and how much
more important?” The pair involved in a comparison is indicated by the (row, column) headings for
the cell into which the judgment is placed. The judgment indicates how much more important the
row element is than the column element in the opinion of the judge. For example, the value of 7 in
the (Consumption, Exports) cell in appendix Table 5 means Consumption is considered to be very
strongly more important than Exports in determining the time to recovery. If instead the judgment
was that Exports were very strongly more important than Consumption, the inverse value of 1/7
would be entered.
Entering the derived priorities in the supermatrix
The priorities in appendix Table 5, determined by computing the principal eigenvector of the
matrix of judgments, are: 0.355, 0.025, 0.053, 0.104, 0.208, 0.227, 0.029. The priorities may be
interpreted as meaning that Consumption at .355 and Monetary Policy at .227 are the most
important Aggregate Demand Factors. Put another way, Consumption is 35.5% of what drives an
economic recovery among aggregate demand factors while Net Exports is the least important at
2.5%. These priorities are entered into the supermatrix in appendix Table 3-a under the Aggregate
Demand column heading.
Posing the question when comparing time periods with respect to criteria
The time periods are compared according to which is the more likely period for recovery due
to the influence of the factor with respect to which the comparisons are made. In Figure 1 the Fiscal
Policy criterion in the Aggregate Demand Factors cluster has the subfactors (or subcriteria): Tax
Policy and Government Expenditure, so the time periods are pairwise compared with respect to
each of these two subfactors rather than directly with respect to the factor Fiscal Policy. The time
periods are compared directly with respect to the factors in the cluster that have no subfactors such
as Consumption. The question posed, for example, would be: Is the Six Months time period or the
Twelve Months time period more likely to be the turnaround time because of Consumption and how
strongly more? Here Six Months is not more likely than Twelve Months; it is the other way around
so a 1/3 is entered in appendix Table 8 in the (Six Months, Twelve Months) cell.
~ 10 ~
Review of Economics & Finance, Volume 5, Issue 2
Posing the question when comparing the primary factors related to time periods
The final type of comparison arises because of the links back from the time periods to the
primary factors. In this case the question posed, for example, is: For the turnaround to occur in the
Six Months time period which would have greater influence, and how strongly greater, the primary
factor Aggregate Demand or Aggregate Supply? The judgment is that Aggregate Demand would be
very strongly dominant over Aggregate Supply for the turnaround to occur in six months, so a 7 is
entered in the (Aggregate Demand, Aggregate Supply) cell in Table 9.
Computing the limit supermatrix
Overall priorities for the alternative time periods are obtained by raising the supermatrix to
powers until it converges to the limit supermatrix shown in Tables 4-a to 4-c. In the limit
supermatrix all the columns are the same. The raw values for the alternative time periods are the
same in every column of the limit supermatrix. These four values are shown in column 2 of
appendix Table 2 below. They are normalized by dividing each by their sum to obtain the priorities
for the time periods shown in the third column. These priorities may be interpreted as the likelihood
of a turnaround occurring during these time periods.
Table 2. Likelihood of a turnaround during a time period
Time period
Raw values
Likelihood of a turnaround during
(from the Limit
time period
Supermatrix)
(Normalize raw values to 1)
Six Months
Twelve Months
Twenty Four Months
Thirty Six Months
0.031
0.063
0.095
0.128
0.097
0.200
0.300
0.404
Computing the expected number of months until the turnaround
To compute the expected time to the turnaround, as is traditionally done in economics,
multiply the mid-point of each time period by the likelihood of the turnaround occurring during that
time period. For example, the Six Months time period runs from 0 to 6 months so its midpoint is 3
months. The Twelve Months time period runs from 6 to 12 months so its midpoint falls at 9 months;
the Twenty Four Months time period runs from 12 to 24 months, so its midpoint falls at 18 months,
and the Thirty Six Months time period runs from 24 to 36, so its midpoint is at 30. Using an
expected value calculation, in months, from Dec.2008: Expected turnaround = 3 × 0.0967 + 9 ×
0.1997 + 18 × 0.3001 + 30 × 0.4035 = 19.5898
The exercise was completed in early December 2008, so 19.59 means the turnaround is
expected to occur in about nineteen and a half months, or around mid July in 2010. We are grateful
to Elsevier Ltd for giving us permission to publish this version of our original paper.
~ 11 ~
ISSNs: 1923-7529; 1923-8401 © 2015 Academic Research Centre of Canada
Table 3-a. Supermatrix that contains the weights derived from the pairwise comparison matrices
0Primary Factors
0 Primary
Factors
Demand
Factors
2 Aggregate
2Aggregate Supply Factors
2Aggreg.
3Global
1Con-
2Net
3Invest-
4Conf-
5Fiscal
6Monetary
7Expected
1Labor
2Natural
3Expec-
Demand
Supply
Finan'l
sump.
Exports
ment
idence
Policy
Policy
Inflation
Costs
Resource $
tations
1 Aggregate Demand
0
0
0
0
0
0
0
0
0
0
0
0
0
2 Aggregate Supply
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1 Consumption
0.355
0
0
0
0
0
0
0
0
0
0
0
0
2 Net Exports
0.025
0
0
0
0
0
0
0
0
0
0
0
0
3 Investment
0.053
0
0
0
0
0
0
0
0
0
0
0
0
4 Confidence
0.104
0
0
0
0
0
0
0
0
0
0
0
0
5 Fiscal Policy
0.208
0
0
0
0
0
0
0
0
0
0
0
0
6 Monetary Policy
0.227
0
0
0
0
0
0
0
0
0
0
0
0
7 Expected Inflation
0.029
0
0
0
0
0
0
0
0
0
0
0
0
1 Labor Costs
0.157
0.157
0
0
0
0
0
0
0
0
0
0
0
3 Global Fnc'l Con.
1 Aggregate
1Aggregate Demand Factors
1Aggreg.
Supply
2 Natural Res. Costs
0.594
0.594
0
0
0
0
0
0
0
0
0
0
0
Factors
3 Expectations
0.249
0.249
0
0
0
0
0
0
0
0
0
0
0
1 Major Int'l Polit.Rel.
0
0
0.033
0
0
0
0
0
0
0
0
0
0
2 Global Fnc'l Integ.
0
0
0.271
0
0
0
0
0
0
0
0
0
0
3 Mortgage Crisis
0
0
0.423
0
0
0
0
0
0
0
0
0
0
4 Expect. Oil Prices
0
0
0.167
0
0
0
0
0
0
0
0
0
0
5 Future Value of $
0
0
0.107
0
0
0
0
0
0
0
0
0
0
1 Tax Policy
0
0
0
0
0
0
0
0.167
0
0
0
0
0
2 Gov't Expenditure
0
0
0
0
0
0
0
0.833
0
0
0
0
0
1 Housing Prices
0
0
0
0
0
0
0
0
0
0
0
0
0
2 Mortgage-bkd Sec.
0
0
0
0
0
0
0
0
0
0
0
0
0
3 Credit Def. Swaps
0
0
0
0
0
0
0
0
0
0
0
0
0
4 Gov't Owner. Inter.
0
0
0
0
0
0
0
0
0
0
0
0
0
5 Financ'l Reporting
0
0
0
0
0
0
0
0
0
0
0
0
0
1Six months
0
0
0
0.051
0.083
0.078
0.09
0
0.19
0.09
0.25
0.078
0.057
2Twelve months
0
0
0
0.083
0.083
0.078
0.149
0
0.445
0.149
0.25
0.134
0.109
3 Global
Financial
Context
11 Fiscal Pol.
Issues
31 Mortgage
Crisis
Issues
4 Alternatives
3Twenty four months
0
0
0
0.307
0.417
0.305
0.223
0
0.258
0.223
0.25
0.323
0.377
4Thirty six months
0
0
0
0.559
0.417
0.538
0.538
0
0.108
0.538
0.25
0.464
0.457
~ 12 ~
Review of Economics & Finance, Volume 5, Issue 2
Table 3-b. Supermatrix (Cont’d)
3Global Financial Contexts
11Fiscal Pol. Issues
31Mortgage Crisis Issues
1Int'l Pol.
2Global
3Mortgage
4Expect'ns
5Future
1Tax
2Gov't
1Uncertainty
2Uncertainty
3Cred. Def.
4Gov't
5Financ'l
Rel's
Financ'l
Crisis
Oil Prices
Value $
Policy
Expend.
Housing $
Mort. Sec.
Swaps
Own.&Int.
Reporting
0 Primary
1 Aggregate Demand
0
0
0
0
0
0
0
0
0
0
0
0
Factors
2 Aggregate Supply
0
0
0
0
0
0
0
0
0
0
0
0
3 Global Fnc'l Con.
0
0
0
0
0
0
0
0
0
0
0
0
1 Aggregate
1 Consumption
0
0
0
0
0
0
0
0
0
0
0
0
Demand
2 Net Exports
0
0
0
0
0
0
0
0
0
0
0
0
Factors
3 Investment
0
0
0
0
0
0
0
0
0
0
0
0
4 Confidence
0
0
0
0
0
0
0
0
0
0
0
0
5 Fiscal Policy
0
0
0
0
0
0
0
0
0
0
0
0
6 Monetary Policy
0
0
0
0
0
0
0
0
0
0
0
0
7 Expected Inflation
0
0
0
0
0
0
0
0
0
0
0
0
1 Labor Costs
0
0
0
0
0
0
0
0
0
0
0
0
2 Natural Resource $
0
0
0
0
0
0
0
0
0
0
0
0
0
2 Aggregate
Supply
Factors
3 Expectations
0
0
0
0
0
0
0
0
0
0
0
1 Major Int'l Polit.Rel.
0
0
0
0
0
0
0
0
0
0
0
0
Financial
2 Global Fnc'l Integ.
0
0
0
0
0
0
0
0
0
0
0
0
Context
3 Mortgage Crisis
0
0
0
0
0
0
0
0
0
0
0
0
4 Expect. Oil Prices
0
0
0
0
0
0
0
0
0
0
0
0
3 Global
11 Fiscal Pol.
Issues
31 Mortgage
5 Future Value of $
0
0
0
0
0
0
0
0
0
0
0
0
1 Tax Policy
0
0
0
0
0
0
0
0
0
0
0
0
2 Gov't Expenditure
0
0
0
0
0
0
0
0
0
0
0
0
1 Housing Prices
0
0
0.3804
0
0
0
0
0
0
0
0
0
Crisis
2 Mortgage-bkd Sec.
0
0
0.2482
0
0
0
0
0
0
0
0
0
Issues
3 Credit Def. Swaps
0
0
0.2506
0
0
0
0
0
0
0
0
0
4 Gov't Owner. Inter.
0
0
0.0444
0
0
0
0
0
0
0
0
0
5 Financ'l Reporting
0
0
0.0764
0
0
0
0
0
0
0
0
0
1Six months
0.0553
0.0553
0
0.0720
0.0720
0.0666
0.1045
0.0591
0.0734
0.0734
0.0734
0.0734
2Twelve months
0.1175
0.1175
0
0.1429
0.1429
0.1416
0.4717
0.1760
0.1728
0.1728
0.1728
0.1728
3Twenty four months
0.2622
0.2622
0
0.3664
0.3664
0.2523
0.2741
0.2888
0.3769
0.3769
0.3769
0.3769
4Thirty six months
0.5650
0.5650
0
0.4188
0.4188
0.5395
0.1498
0.4762
0.3769
0.3769
0.3769
0.3769
4 Alternatives
~ 13 ~
ISSNs: 1923-7529; 1923-8401 © 2015 Academic Research Centre of Canada
Table 3-c. Supermatrix (Cont’d)
4Alternatives
1Six
2Twelve
months
months
3Twenty four
months
4Thirty six
months
0 Primary
1 Aggregate Demand
0.7854
0.7854
0.4545
0.4545
Factors
2 Aggregate Supply
0.1488
0.1488
0.4545
0.4545
3 Global Fnc'l Con.
0.0658
0.0658
0.0909
0.0909
1 Aggregate
1 Consumption
0
0
0
0
Demand
2 Net Exports
0
0
0
0
Factors
3 Investment
0
0
0
0
4 Confidence
0
0
0
0
5 Fiscal Policy
0
0
0
0
6 Monetary Policy
0
0
0
0
7 Expected Inflation
0
0
0
0
1 Labor Costs
0
0
0
0
2 Aggregate
Supply
2 Natural Resource $
0
0
0
0
Factors
3 Expectations
0
0
0
0
3 Global
1 Major Int'l Polit.Rel.
0
0
0
0
Financial
2 Global Fnc'l Integ.
0
0
0
0
Context
3 Mortgage Crisis
0
0
0
0
4 Expect. Oil Prices
0
0
0
0
5 Future Value of $
0
0
0
0
11 Fiscal Pol.
1 Tax Policy
0
0
0
0
Issues
2 Gov't Expenditure
0
0
0
0
31 Mortgage
1 Housing Prices
0
0
0
0
Crisis
2 Mortgage-bkd Sec.
0
0
0
0
Issues
3 Credit Def. Swaps
0
0
0
0
4 Gov't Owner. Inter.
0
0
0
0
5 Financ'l Reporting
0
0
0
0
1Six months
0
0
0
0
2Twelve months
0
0
0
0
3Twenty four months
4Thirty six months
0
0
0
0
0
0
0
0
4 Alternatives
~ 14 ~
Review of Economics & Finance, Volume 5, Issue 2
Table 4-a. Limit Supermatrix (for this model all the columns have the same values)
0Primary Factors
1Aggregate Demand Factors
2Aggregate Supply Factors
1Aggreg.
2Aggreg.
3Global
1Con-
2Net
3Invest-
4Conf-
5Fiscal
6Mone-
7Expected
1Labor
2Natural
3Expect-
Demand
Supply
Finan'l
sump.
Exports
ment
idence
Policy
tary Pol.
Inflation
Costs
Resource $
ations
0 Primary
1 Aggregate Demand
0.1755
0.1755
0.1755
0.1755
0.1755
0.1755
0.1755
0.1755
0.1755
0.1755
0.1755
0.1755
0.1755
Factors
2 Aggregate Supply
0.1155
0.1155
0.1155
0.1155
0.1155
0.1155
0.1155
0.1155
0.1155
0.1155
0.1155
0.1155
0.1155
3 Global Fnc'l Con.
0.0265
0.0265
0.0265
0.0265
0.0265
0.0265
0.0265
0.0265
0.0265
0.0265
0.0265
0.0265
0.0265
1 Aggregate
1 Consumption
0.0623
0.0623
0.0623
0.0623
0.0623
0.0623
0.0623
0.0623
0.0623
0.0623
0.0623
0.0623
0.0623
Demand
2 Net Exports
0.0044
0.0044
0.0044
0.0044
0.0044
0.0044
0.0044
0.0044
0.0044
0.0044
0.0044
0.0044
0.0044
Factors
3 Investment
0.0092
0.0092
0.0092
0.0092
0.0092
0.0092
0.0092
0.0092
0.0092
0.0092
0.0092
0.0092
0.0092
4 Confidence
0.0183
0.0183
0.0183
0.0183
0.0183
0.0183
0.0183
0.0183
0.0183
0.0183
0.0183
0.0183
0.0183
5 Fiscal Policy
0.0364
0.0364
0.0364
0.0364
0.0364
0.0364
0.0364
0.0364
0.0364
0.0364
0.0364
0.0364
0.0364
6 Monetary Policy
0.0398
0.0398
0.0398
0.0398
0.0398
0.0398
0.0398
0.0398
0.0398
0.0398
0.0398
0.0398
0.0398
2 Aggregate
7 Expected Inflation
0.0051
0.0051
0.0051
0.0051
0.0051
0.0051
0.0051
0.0051
0.0051
0.0051
0.0051
0.0051
0.0051
1 Labor Costs
0.0181
0.0181
0.0181
0.0181
0.0181
0.0181
0.0181
0.0181
0.0181
0.0181
0.0181
0.0181
0.0181
Supply
2 Natural Resource $
0.0686
0.0686
0.0686
0.0686
0.0686
0.0686
0.0686
0.0686
0.0686
0.0686
0.0686
0.0686
0.0686
Factors
3 Expectations
0.0288
0.0288
0.0288
0.0288
0.0288
0.0288
0.0288
0.0288
0.0288
0.0288
0.0288
0.0288
0.0288
3 Global
1 Major Int'l Polit.Rel.
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
Financial
2 Global Fnc'l Integ.
0.0072
0.0072
0.0072
0.0072
0.0072
0.0072
0.0072
0.0072
0.0072
0.0072
0.0072
0.0072
0.0072
Context
3 Mortgage Crisis
0.0112
0.0112
0.0112
0.0112
0.0112
0.0112
0.0112
0.0112
0.0112
0.0112
0.0112
0.0112
0.0112
4 Expect. Oil Prices
0.0044
0.0044
0.0044
0.0044
0.0044
0.0044
0.0044
0.0044
0.0044
0.0044
0.0044
0.0044
0.0044
5 Future Value of $
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
11 Fiscal Pol.
Issues
31 Mortgage
1 Tax Policy
0.0061
0.0061
0.0061
0.0061
0.0061
0.0061
0.0061
0.0061
0.0061
0.0061
0.0061
0.0061
0.0061
2 Gov't Expenditure
0.0303
0.0303
0.0303
0.0303
0.0303
0.0303
0.0303
0.0303
0.0303
0.0303
0.0303
0.0303
0.0303
1 Housing Prices
0.0043
0.0043
0.0043
0.0043
0.0043
0.0043
0.0043
0.0043
0.0043
0.0043
0.0043
0.0043
0.0043
Crisis
2 Mortgage-bkd Sec.
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
Issues
3 Credit Def. Swaps
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
4 Gov't Owner. Inter.
0.0005
0.0005
0.0005
0.0005
0.0005
0.0005
0.0005
0.0005
0.0005
0.0005
0.0005
0.0005
0.0005
5 Financ'l Reporting
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
1Six months
0.0307
0.0307
0.0307
0.0307
0.0307
0.0307
0.0307
0.0307
0.0307
0.0307
0.0307
0.0307
0.0307
2Twelve months
0.0634
0.0634
0.0634
0.0634
0.0634
0.0634
0.0634
0.0634
0.0634
0.0634
0.0634
0.0634
0.0634
3Twenty four months
0.0953
0.0953
0.0953
0.0953
0.0953
0.0953
0.0953
0.0953
0.0953
0.0953
0.0953
0.0953
0.0953
4Thirty six months
0.1281
0.1281
0.1281
0.1281
0.1281
0.1281
0.1281
0.1281
0.1281
0.1281
0.1281
0.1281
0.1281
4 Alternatives
~ 15 ~
ISSNs: 1923-7529; 1923-8401 © 2015 Academic Research Centre of Canada
Table 4-b. Limit Supermatrix (Cont’d)
3Global Financial Contexts
1 Int'l
2Global
3Mortgage
Pol.
Rel'ns
Fnc'l
Crisis
Int.
4Expect'ns
5Future
Oil Prices
Value $
11Fiscal Policy Issues
1Tax
2Gov't
Policy
Expend.
31Mortgage Crisis Issues
1Uncertainty
2Mort.
Bkd
Housing $
Securities
3Role
Credit
Default
Swps
4Gov't
5Financ'l
Own.&Int.
Reporting
0 Primary
1 Aggregate Demand
0.1755
0.1755
0.1755
0.1755
0.1755
0.1755
0.1755
0.1755
0.1755
0.1755
0.1755
0.1755
Factors
2 Aggregate Supply
0.1155
0.1155
0.1155
0.1155
0.1155
0.1155
0.1155
0.1155
0.1155
0.1155
0.1155
0.1155
3 Global Fnc'l Con.
0.0265
0.0265
0.0265
0.0265
0.0265
0.0265
0.0265
0.0265
0.0265
0.0265
0.0265
0.0265
1 Consumption
0.0623
0.0623
0.0623
0.0623
0.0623
0.0623
0.0623
0.0623
0.0623
0.0623
0.0623
0.0623
1 Aggregate
Demand
2 Net Exports
0.0044
0.0044
0.0044
0.0044
0.0044
0.0044
0.0044
0.0044
0.0044
0.0044
0.0044
0.0044
Factors
3 Investment
0.0092
0.0092
0.0092
0.0092
0.0092
0.0092
0.0092
0.0092
0.0092
0.0092
0.0092
0.0092
4 Confidence
0.0183
0.0183
0.0183
0.0183
0.0183
0.0183
0.0183
0.0183
0.0183
0.0183
0.0183
0.0183
5 Fiscal Policy
0.0364
0.0364
0.0364
0.0364
0.0364
0.0364
0.0364
0.0364
0.0364
0.0364
0.0364
0.0364
6 Monetary Policy
0.0398
0.0398
0.0398
0.0398
0.0398
0.0398
0.0398
0.0398
0.0398
0.0398
0.0398
0.0398
7 Expected Inflation
0.0051
0.0051
0.0051
0.0051
0.0051
0.0051
0.0051
0.0051
0.0051
0.0051
0.0051
0.0051
1 Labor Costs
0.0181
0.0181
0.0181
0.0181
0.0181
0.0181
0.0181
0.0181
0.0181
0.0181
0.0181
0.0181
Supply
2 Natural Resource $
0.0686
0.0686
0.0686
0.0686
0.0686
0.0686
0.0686
0.0686
0.0686
0.0686
0.0686
0.0686
Factors
3 Expectations
0.0288
0.0288
0.0288
0.0288
0.0288
0.0288
0.0288
0.0288
0.0288
0.0288
0.0288
0.0288
2 Aggregate
3 Global
1 Major Int'l Polit.Rel.
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
Financial
2 Global Fnc'l Integ.
0.0072
0.0072
0.0072
0.0072
0.0072
0.0072
0.0072
0.0072
0.0072
0.0072
0.0072
0.0072
Context
3 Mortgage Crisis
0.0112
0.0112
0.0112
0.0112
0.0112
0.0112
0.0112
0.0112
0.0112
0.0112
0.0112
0.0112
4 Expect. Oil Prices
0.0044
0.0044
0.0044
0.0044
0.0044
0.0044
0.0044
0.0044
0.0044
0.0044
0.0044
0.0044
5 Future Value of $
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
1 Tax Policy
0.0061
0.0061
0.0061
0.0061
0.0061
0.0061
0.0061
0.0061
0.0061
0.0061
0.0061
0.0061
2 Gov't Expenditure
0.0303
0.0303
0.0303
0.0303
0.0303
0.0303
0.0303
0.0303
0.0303
0.0303
0.0303
0.0303
1 Housing Prices
0.0043
0.0043
0.0043
0.0043
0.0043
0.0043
0.0043
0.0043
0.0043
0.0043
0.0043
0.0043
2 Mortgage-bkd Sec.
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
11 Fiscal Pol.
Issues
31 Mortgage
Crisis
Issues
4 Alternatives
3 Credit Def. Swaps
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
0.0028
4 Gov't Owner. Inter.
0.0005
0.0005
0.0005
0.0005
0.0005
0.0005
0.0005
0.0005
0.0005
0.0005
0.0005
0.0005
5 Financ'l Reporting
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
0.0009
1Six months
0.0307
0.0307
0.0307
0.0307
0.0307
0.0307
0.0307
0.0307
0.0307
0.0307
0.0307
0.0307
2Twelve months
0.0634
0.0634
0.0634
0.0634
0.0634
0.0634
0.0634
0.0634
0.0634
0.0634
0.0634
0.0634
3Twenty four months
0.0953
0.0953
0.0953
0.0953
0.0953
0.0953
0.0953
0.0953
0.0953
0.0953
0.0953
0.0953
4Thirty six months
0.1281
0.1281
0.1281
0.1281
0.1281
0.1281
0.1281
0.1281
0.1281
0.1281
0.1281
0.1281
~ 16 ~
Review of Economics & Finance, Volume 5, Issue 2
Table 4-c. Limit Supermatrix (Cont’d)
4Alternatives
1Six
2Twelve
months
months
3Twenty
four
months
4Thirty
six
months
0 Primary
1 Aggregate Demand
0.1755
0.1755
0.1755
0.1755
Factors
2 Aggregate Supply
0.1155
0.1155
0.1155
0.1155
3 Global Fnc'l Con.
0.0265
0.0265
0.0265
0.0265
1 Consumption
0.0623
0.0623
0.0623
0.0623
Demand
2 Net Exports
0.0044
0.0044
0.0044
0.0044
Factors
3 Investment
0.0092
0.0092
0.0092
0.0092
4 Confidence
0.0183
0.0183
0.0183
0.0183
5 Fiscal Policy
0.0364
0.0364
0.0364
0.0364
6 Monetary Policy
0.0398
0.0398
0.0398
0.0398
7 Expected Inflation
0.0051
0.0051
0.0051
0.0051
1 Aggregate
2 Aggregate
1 Labor Costs
0.0181
0.0181
0.0181
0.0181
Supply
2 Natural Resource $
0.0686
0.0686
0.0686
0.0686
Factors
3 Expectations
0.0288
0.0288
0.0288
0.0288
1 Major Int'l Polit.Rel.
0.0009
0.0009
0.0009
0.0009
Financial
2 Global Fnc'l Integ.
0.0072
0.0072
0.0072
0.0072
Context
3 Mortgage Crisis
0.0112
0.0112
0.0112
0.0112
4 Expect. Oil Prices
0.0044
0.0044
0.0044
0.0044
5 Future Value of $
0.0028
0.0028
0.0028
0.0028
1 Tax Policy
0.0061
0.0061
0.0061
0.0061
2 Gov't Expenditure
0.0303
0.0303
0.0303
0.0303
3 Global
11 Fiscal Pol.
Issues
31 Mortgage
1 Housing Prices
0.0043
0.0043
0.0043
0.0043
Crisis
2 Mortgage-bkd Sec.
0.0028
0.0028
0.0028
0.0028
Issues
3 Credit Def. Swaps
0.0028
0.0028
0.0028
0.0028
4 Gov't Owner. Inter.
0.0005
0.0005
0.0005
0.0005
5 Financ'l Reporting
0.0009
0.0009
0.0009
0.0009
1Six months
0.0307
0.0307
0.0307
0.0307
2Twelve months
0.0634
0.0634
0.0634
0.0634
3Twenty four months
0.0953
0.0953
0.0953
0.0953
4Thirty six months
0.1281
0.1281
0.1281
0.1281
4 Alternatives
Pairwise Comparisons of the Elements of the Model
A set of priorities or weights was obtained from each of 29
pairwise comparison matrices as its principal eigenvector. These derived
priorities were placed into the supermatrix in the appropriate column.
For example, appendix Table 5 below gives the priorities or weights for
the Aggregate Demand Main Factor, so they are placed in the
supermatrix in appendix Table 3-a in the first column under “Primary
Factors Aggregate Demand” beginning in the fourth row.
The judgments and resulting weights are also shown in appendix
Tables 6 and 7 for the Aggregate Supply Main Factor and for the Global
Financial Context Main Factor. Also shown in appendix Table 8 are the
judgments and weights for the time alternatives relating to Consumption.
Finally, appendix Table 9 shows the judgments and weights for the
Primary Factors (i.e., Aggregate Demand, Aggregate Supply, and the
Global Financial Context) with respect to the 6-month time period. The
other judgments and weights for the other elements in the model can be
found in tables appended to Blair, Mandelker, Saaty and Whitaker
(2010).
~ 17 ~
Review of Economics & Finance
Submitted on 03/11/2014
Article ID: 1923-7529-2015-02-01-18
Andrew R. Blair, Gershon Mandelker, Thomas L. Saaty, and Rozann Whitaker
Table 5. The judgments for the subfactors of the aggregate demand main factor
1Consum-ption
2Net Exports
3Invest-ment
4Confid-ence
5Fiscal Policy
6Monetary Policy
7Expected Inflation
1Consumption
1
1/7
1/7
1/5
1/3
1/2
1/7
2Exports
7
1
5
7
8
7
1
3Invest
-ment
7
1/5
1
3
5
6
1/2
4Confidence
5
1/7
1/3
1
4
4
1/6
5Fiscal
Policy
3
1/8
1/5
1/4
1
1
1/5
6Monetary
Policy
2
1/7
1/6
1/4
1
1
1/7
7Expected
Inflation
7
1
2
6
5
7
1
Weights
0.355
0.025
0.053
0.104
0.208
0.227
0.029
Table 6. The judgments for the subfactors of the aggregate supply main factor
Labor Costs
Natural Resources
Expectations
Weights
1
3
2
1/3
1
1/3
1/2
3
1
.157
.594
.249
Labor Costs
Natural Resources
Expectations
Table 7. Judgments for the subfactors of the global financial context main factor
Major Int’l Political
Relationships
Global Financial
Integration
Mortgage Crisis Issues
Expectations of Future
Oil Prices
Future Value of the
Dollar
Major
International
Political
Relationships
Global
Financial
Integration
Mortgage
Crisis
Issues
Expectations
of Future Oil
Prices
1
1/7
1/7
1/6
1/6
.033
7
1
1
2
2
.271
7
1
1
5
5
.423
6
1/2
1/5
1
3
.167
6
1/2
1/5
1/3
1
.107
Future
Value of
the Dollar
Weights
Table 8. The judgments for the alternatives with respect to consumption
6 Months
6 Months
12Months
24 Months
36 Months
1
3
5
7
12 Months
24 Months
1/3
1
7
7
1/5
1/7
1
3
36 Months
1/7
1/7
1/3
1
Weights
0.051
0.083
0.307
0.559
Table 9. Judgments for the primary factors with respect to 6 month time period
Aggregate Demand
Aggregate Supply
Geopolitical
Aggregate
Demand
1
1/7
1/9
Aggregate
Supply
7
1
1/3
~ 18 ~
Global Financial
Context
9
3
1
Weights
0.785
0.149
0.066