Discussion of Discussion of Irvine and SchuhIrvine and Schuh

Robert J. GordonRobert J. GordonNorthwestern University and Northwestern University and

NBERNBERFRB San FranciscoFRB San FranciscoNovember 3, 2007November 3, 2007

This Paper is Novel and This Paper is Novel and ImportantImportant

The Great Moderation is not caused byThe Great Moderation is not caused by ““Good Luck”Good Luck” Better Monetary PolicyBetter Monetary Policy

Rather, 80% of reduced volatility is explained Rather, 80% of reduced volatility is explained by changes in the structural relationships by changes in the structural relationships between industry-sector sales and inventory between industry-sector sales and inventory investmentinvestment

We only need to look at manufacturing and We only need to look at manufacturing and tradetrade Can neglect such previous “usual suspects” as Can neglect such previous “usual suspects” as

military spending and residential constructionmilitary spending and residential construction

This Discussion, like Gaul, This Discussion, like Gaul, is Divided into Three Partsis Divided into Three Parts

The first part summarizes what I The first part summarizes what I thought about the Great Moderation thought about the Great Moderation before reading this paperbefore reading this paper

The second part summarizes the most The second part summarizes the most important results of the authorsimportant results of the authors

The third part ponders the significance The third part ponders the significance of the paper’s results: by how much of the paper’s results: by how much do I need to change my previous do I need to change my previous interpretation of the Great Moderationinterpretation of the Great Moderation

My Interpretation of the My Interpretation of the Great ModerationGreat Moderation

This is from NBER WP 11777 in This is from NBER WP 11777 in November 2005November 2005

Published in an obscure conference Published in an obscure conference volume of the Reserve Bank of Australia, volume of the Reserve Bank of Australia, where the volume is devoted to exactly where the volume is devoted to exactly the same topic as the current SF the same topic as the current SF conference.conference.

Some of the papers in that conference Some of the papers in that conference volume are worth looking up, not just volume are worth looking up, not just minemine

Stabilization before and Stabilization before and after 1984after 1984

ShocksShocks Demand shocksDemand shocks

Federal government now the culprit not the Federal government now the culprit not the salvationsalvation

Inventory managementInventory management Financial Market Deregulation stabilized residential Financial Market Deregulation stabilized residential

housinghousing Supply shocksSupply shocks

Improved monetary policyImproved monetary policy Of Lesser ImportanceOf Lesser Importance

Shifts in shares to servicesShifts in shares to services

Inflation vs. Output Inflation vs. Output Volatility:Volatility:

20-quarter rolling standard 20-quarter rolling standard deviationdeviation

of 4-quarter growth ratesof 4-quarter growth rates

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005

Real GDP Growth Volatility

Inflation Volatility

Summary of inflation Summary of inflation volatilityvolatility

vs. real GDP volatility (20 vs. real GDP volatility (20 qtr st dev)qtr st dev)1952-721952-72 1973-871973-87 1988-1988-

20052005

Real GDPReal GDP 2.692.69 2.872.87 1.251.25

GDP DeflatorGDP Deflator 1.111.11 1.671.67 0.480.48

Demand Side: Demand Side: Decomposition of GDP Decomposition of GDP

Contributions by 11 SectorsContributions by 11 SectorsStandard Deviations of 4-quarter Moving Standard Deviations of 4-quarter Moving

Averages of a Sector’s Contribution to Averages of a Sector’s Contribution to ΔΔ Real GDPReal GDP

1950-831950-83 1984-20051984-2005 DiffDiff %%Real GDPReal GDP 3.143.141.611.61-1.53-1.53 100100Omit RSOmit RS 2.782.781.441.44-1.34-1.34 88 88Omit IIOmit II 2.442.441.331.33-1.11-1.11 73 73Omit Fed GovtOmit Fed Govt 3.183.181.611.61-1.57-1.57 103103Omit All 3Omit All 3 1.931.931.191.19-0.74-0.74 48 48

This Raises my First This Raises my First QuestionQuestion

Inventory Change Accounts for 27%Inventory Change Accounts for 27% Inventory Changes, Residential Inventory Changes, Residential

Structures, and Federal Govt Account Structures, and Federal Govt Account for 52%for 52%

How Can a Paper That Covers only How Can a Paper That Covers only Manufacturing and Trade Account for Manufacturing and Trade Account for most of the reduction in volatility?most of the reduction in volatility?

Consider the Possibility that the Consider the Possibility that the Shocks Feeding into their Structural Shocks Feeding into their Structural Mechanism have Reduced VolatilityMechanism have Reduced Volatility

Contrast their HAVAR with Contrast their HAVAR with my Three Equation Modelmy Three Equation Model

based on Stock-Watson based on Stock-Watson Combines my “mainstream” or “triangle” Combines my “mainstream” or “triangle”

approach to explaining inflationapproach to explaining inflation InertiaInertia Demand through output or U gapDemand through output or U gap Specific supply shocksSpecific supply shocks

““Taylor Rule” equation for Fed Funds rateTaylor Rule” equation for Fed Funds rate Coefficients allowed to change, 1979 and 1990Coefficients allowed to change, 1979 and 1990

Output gap equation with feedback from interest Output gap equation with feedback from interest rate changesrate changes

Main difference from Stock-Watson (2002,2003) is Main difference from Stock-Watson (2002,2003) is the use of specific supply shock variables instead the use of specific supply shock variables instead of stuffing them into the error termof stuffing them into the error term

The Supply Shocks are The Supply Shocks are ImportantImportant

and have been Neglected and have been Neglected HereHere Everything is expressed as a relative rate of Everything is expressed as a relative rate of

change. A zero value means no impact on change. A zero value means no impact on aggregate inflationaggregate inflation

The list of fourThe list of four Food-energy effect (difference headline vs. core inf)Food-energy effect (difference headline vs. core inf) Changes in relative price of importsChanges in relative price of imports Changes in the productivity growth trendChanges in the productivity growth trend Nixon-era price controls, “on-off” dummies adding to Nixon-era price controls, “on-off” dummies adding to

zerozero Next slide shows effect of suppressing all the Next slide shows effect of suppressing all the

supply shocks; all that’s left is effect of LDV and supply shocks; all that’s left is effect of LDV and Ugap.Ugap.

The Dramatic Effect of The Dramatic Effect of Supply ShocksSupply Shocks

-6

-4

-2

0

2

4

6

8

10

12

1960 1965 1970 1975 1980 1985 1990 1995 2000

Predicted Inflation w ith Actual Shocks, 1965-2004

Predicted Inflation w ith ShocksSuppressed, 1965-2004

Results from the No-SSResults from the No-SSSimulationSimulation

The no-SS simulation is driven entirely The no-SS simulation is driven entirely by the LDV and the current and 4 lags by the LDV and the current and 4 lags on the unemployment gapon the unemployment gap

No difference until 1973No difference until 1973 Without SS, inflation goes negative Without SS, inflation goes negative

after 1982. But Volcker inflation-after 1982. But Volcker inflation-fighting would have been unnecessary fighting would have been unnecessary without SSwithout SS

Difference narrows in late 1990s, why?Difference narrows in late 1990s, why?

Full Model Simulations: Full Model Simulations: Here is InflationHere is Inflation

-2

0

2

4

6

8

10

12

14

1965:01 1970:01 1975:01 1980:01 1985:01 1990:01 1995:01 2000:01

All Shocks

No Interest Error

No Output ErrorNo Shocks

No Supply Shocks

The Basic Conclusion of the The Basic Conclusion of the Paper:Paper:

The Output Gap The Output Gap SimulationsSimulations

-12

-10

-8

-6

-4

-2

0

2

4

6

8

1965:01 1970:01 1975:01 1980:01 1985:01 1990:01 1995:01 2000:01

All Shocks

No Output Error

No Interest Error

No Shocks

No Supply Shocks

ConclusionsConclusions

Demand and Supply Shocks both Demand and Supply Shocks both MatteredMattered The Major Demand Shocks were Military The Major Demand Shocks were Military

Spending, Financial Institutions that Spending, Financial Institutions that Destabilized Residential Investment, and Destabilized Residential Investment, and Primitive Inventory ManagementPrimitive Inventory Management

The Major Supply Shocks were Import The Major Supply Shocks were Import Prices (and Flexible Exchange Rates), Prices (and Flexible Exchange Rates), Food-Oil Prices, Productivity Trend, and Food-Oil Prices, Productivity Trend, and Nixon ControlsNixon Controls

Full-Model SimulationsFull-Model Simulations

Comparing 1965-83 with 1984-2004Comparing 1965-83 with 1984-2004 Inflation VolatilityInflation Volatility

Reversal of SS Accounts for 80%, Reversal of SS Accounts for 80%, Output Error 20%Output Error 20%

Output VolatilityOutput Volatility St Dev 2/3 explained by OE in both St Dev 2/3 explained by OE in both

periodsperiods SS contributed about 1/3 in first periodSS contributed about 1/3 in first period

Monetary PolicyMonetary Policy Big Surprise, Greenspan = BurnsBig Surprise, Greenspan = Burns Narrow View: Many other changesNarrow View: Many other changes

Credibility Because there was no inflationCredibility Because there was no inflation Would have behaved differently if there had Would have behaved differently if there had

been more inflationbeen more inflation Inflation-Output Gap Tradeoff Lives OnInflation-Output Gap Tradeoff Lives On

Greenspan policies throughout would have Greenspan policies throughout would have delivered 5 points higher inflation post-84delivered 5 points higher inflation post-84

Output benefits only temporaryOutput benefits only temporary

Irvine-Schuh Conclude 80%Irvine-Schuh Conclude 80%Structural Change not Structural Change not

“Good Luck”“Good Luck” What does “Good Luck” Mean?What does “Good Luck” Mean?

Switch of SS from bad to good is indeed Switch of SS from bad to good is indeed “Good Luck”“Good Luck”

But financial deregulation that reduced But financial deregulation that reduced residential construction volatility is policy, residential construction volatility is policy, not good lucknot good luck

Reduced size and volatility of military Reduced size and volatility of military spending is policy, not good luckspending is policy, not good luck

Improved inventory management results Improved inventory management results from technology, so “good luck” is a from technology, so “good luck” is a misnomer alsomisnomer also

Summary of Paper’s Summary of Paper’s ResultsResults

Point of Departure, VARPoint of Departure, VAR 21% of Great Moderation to Structural Change, 79% 21% of Great Moderation to Structural Change, 79%

to “Good Luck”to “Good Luck” Their 3-sector HAVAR attributes 73% to Their 3-sector HAVAR attributes 73% to

Structural Change, only 27% left for Good LuckStructural Change, only 27% left for Good Luck Since Improved Inventory Management is the Since Improved Inventory Management is the

top item on my list, I support the overall theme top item on my list, I support the overall theme of their paperof their paper

Qualification on p. 3: “A single, or even unified Qualification on p. 3: “A single, or even unified explanation, for the Great Moderation may be explanation, for the Great Moderation may be unlikely”unlikely” I agree, because I have already pointed to four I agree, because I have already pointed to four

explanations, not just oneexplanations, not just one

Consider the Auto Consider the Auto IndustryIndustry

My Story, “changed structure” represents reduced My Story, “changed structure” represents reduced macro volatility from other sourcesmacro volatility from other sources

Faced with much reduced sales shocks, firms can Faced with much reduced sales shocks, firms can and did manage inventories betterand did manage inventories better This can account for much of the reduced covariance This can account for much of the reduced covariance

between sales and inventories, between industry j and kbetween sales and inventories, between industry j and k Don’t forget 1970:Q4 GM strike, that huge spike Don’t forget 1970:Q4 GM strike, that huge spike

in Figure 1. Yes, absence of auto strikes and labor in Figure 1. Yes, absence of auto strikes and labor strife is a structural changestrife is a structural change

Good points in auto discussion: Dealers sell Good points in auto discussion: Dealers sell multiple brands, role of imports and exportsmultiple brands, role of imports and exports

Don’t forget Toyota “pull vs. push” as foreign Don’t forget Toyota “pull vs. push” as foreign manufacturers invade US with a different system manufacturers invade US with a different system (Toyota operates with ½ the inventories per (Toyota operates with ½ the inventories per market share point, this week’s WSJ)market share point, this week’s WSJ)

Interpreting this Paper:Interpreting this Paper:Impulse vs. Propagation Impulse vs. Propagation

MechanismsMechanisms By omitting any mention of military spending, By omitting any mention of military spending,

residential construction, or inflation supply residential construction, or inflation supply shocks, they “import unexplained” into their shocks, they “import unexplained” into their analytical structure at least half of the decline analytical structure at least half of the decline in output volatilityin output volatility

All their metrics of reduced volatility are as a All their metrics of reduced volatility are as a percentage of M&T variance, not total percentage of M&T variance, not total economy variance. economy variance.

By the way, why does data analysis extend By the way, why does data analysis extend only to 2001?only to 2001?

Covariance between SalesCovariance between Salesand Inventory Investment and Inventory Investment

(Table 3)(Table 3) What Table 3 shows is a radical decline What Table 3 shows is a radical decline

in the late/early ratios in every rowin the late/early ratios in every row Variances and covariances declined in Variances and covariances declined in

every rowevery row No evidence here for a change in No evidence here for a change in

structure, rather this seems compatible structure, rather this seems compatible with some outside force reducing with some outside force reducing variance in sales which allowed reduction variance in sales which allowed reduction in variance of inventories and in in variance of inventories and in covariancecovariance

Advantages of HAVAR Advantages of HAVAR ModelModel

Any Model that Nests other Models Any Model that Nests other Models is Goodis Good

Can Measure Significance of Implicit Can Measure Significance of Implicit RestrictionsRestrictions

However this works both waysHowever this works both ways My inflation model nests any simple My inflation model nests any simple

VAR approach as in this paperVAR approach as in this paper

The HAVAR inflation The HAVAR inflation equation equation

is nested in mineis nested in mine Like Stock-Watson, the inflation Like Stock-Watson, the inflation

equation depends only on the output equation depends only on the output gap and Fed Funds rategap and Fed Funds rate

All supply shocks are stuffed into the All supply shocks are stuffed into the error termerror term

Short lags on lagged inflationShort lags on lagged inflation Example of the flaws of this approachExample of the flaws of this approach

Consider John Roberts of the FedConsider John Roberts of the Fed Inflation depends on the unemployment Inflation depends on the unemployment

gap and four lags of inflationgap and four lags of inflation

Bias in Size and Drift of Phillips Curve Slope

Figure 9. Roberts Vs. Triangle Unemployment Coefficients on 90 Quarter Rolling Regressions from 1962:Q1 to 1984:Q3

-1

-0.9

-0.8

-0.7

-0.6

-0.5

-0.4

-0.3

-0.2

-0.1

0

1963 1966 1969 1972 1975 1978 1981 1984

Roberts

Triangle

Post-Sample Dynamic Simulations

Figure 8. Predicted and Simulated Values of Inflation from Triangle and Roberts Equations 1962:Q1 to 2006:Q4

0

2

4

6

8

10

12

1963 1967 1971 1975 1979 1983 1987 1991 1995 1999 2003 2007

Inflation

Triangle

Roberts

Back to the HAVAR Back to the HAVAR StructureStructure

Assumption (p. 17) that II does not affect Assumption (p. 17) that II does not affect sales contemporaneously but sales do affect sales contemporaneously but sales do affect IIII

This is violated by the everyday behavior of This is violated by the everyday behavior of the auto industrythe auto industry Overproduction leads to price incentives, interest Overproduction leads to price incentives, interest

rate incentives that directly increase salesrate incentives that directly increase sales Violated every day also in today’s housing Violated every day also in today’s housing

industry, where excess inventories lead to industry, where excess inventories lead to price reductions in order to stimulate salesprice reductions in order to stimulate sales

Also conflicts with bottom p. 18 “inventories Also conflicts with bottom p. 18 “inventories in one sector might plausibly affect sales in in one sector might plausibly affect sales in the other sector”the other sector”

Other Aspects of HAVAROther Aspects of HAVAR

Unlike Stock-Watson and othersUnlike Stock-Watson and others No attempt to portray differences in No attempt to portray differences in

monetary regimesmonetary regimes Other papers with this VAR structure allow Other papers with this VAR structure allow

for shifts of coefficients in the interest rate for shifts of coefficients in the interest rate equation in 1979 and 1987 or 1990equation in 1979 and 1987 or 1990

Discussion of sales persistence in autosDiscussion of sales persistence in autos Not enough discussion of increased price Not enough discussion of increased price

flexibilityflexibility Price incentives and interest rate incentivesPrice incentives and interest rate incentives

General ConclusionGeneral Conclusion

Link with Gambetti-Gali paperLink with Gambetti-Gali paper My interpretation of hours-productivity My interpretation of hours-productivity

correlation combines positive and correlation combines positive and negative correlationnegative correlation

As overall volatility is reduced, the As overall volatility is reduced, the share of positive correlation is reduced share of positive correlation is reduced that that of negative correlation that that of negative correlation increasesincreases

Something like that may be happening Something like that may be happening in the structural dynamics of this paperin the structural dynamics of this paper