Gordon (2005) investigates the causes of the U.S. Great Moderation — the sharp post-1984 decline in the volatility of real GDP growth, the output gap, and inflation. Using a small four-equation macro model (inflation, Taylor rule, IS/output gap, Okun's law) with explicit supply shock variables, it concludes that roughly 80% of pre-1984 inflation volatility is explained by supply shocks (import prices, food-energy, medical care, productivity trend, Nixon controls) and that most output gap volatility reflects unexplained IS shifts rather than monetary policy responses. The paper's most provocative finding is that, after correcting for positive serial correlation, the Greenspan-era (1990–2004) Taylor Rule reaction function is statistically indistinguishable from the pre-Volcker Burns-era function: the apparent inflation-fighting credentials of the Greenspan Fed vanish once the serial correlation problem is addressed.
Break date: Rolling 20-quarter standard deviation (SD) of real GDP growth fell from 2.76% (1952–87Q4) to 1.25% (1988–2005Q1), a 55% drop. Inflation SD fell ~60%. Break coincides with end of 1981–82 recession.
Three sectors account for ~50% of variance decline: Residential investment + inventory investment + Federal government spending explain half of the GDP volatility reduction despite representing only 17%→13% of nominal GDP across the break. Their stabilization is attributed respectively to financial market reform, IT-enabled inventory management, and declining military share of GDP.
Share shifts vs. within-component change: ~80% of reduced output volatility reflects lower volatility within each GDP component; ~20% comes from compositional shifts toward stable consumer services.
Supply shocks dominate inflation: The triangle model inflation equation attributes ~80% of pre-1984 inflation variance to supply shocks (import price changes, food-energy effects, medical care effects, Hodrick-Prescott (HP)-filtered productivity trend acceleration, Nixon controls); ~20% to IS shifts via the output error.
Output gap variance: IS shifts dominant: The residual in the output-gap equation ("IS shifts") explains more than two-thirds of output gap variance in both sub-periods. Supply shocks explain ~40% of pre-1984 output gap volatility but approximately zero in the post-1984 period (beneficial supply shocks offset adverse ones).
Greenspan ≈ Burns after AR(1) correction: Without serial correlation correction, Greenspan-era (1990–2004) Taylor Rule inflation coefficient = 1.43 (inflation-fighting); after feasible generalized least squares (FGLS) AR(1) correction, drops to 0.57 (inflation-accommodating), essentially identical to the Burns-era coefficient (0.57). Output gap response also falls from 0.95 to 0.60. Only the Volcker era (1979–90) coefficient (1.46) is genuinely different.
Sacrifice ratio: Actual 1981–85 sacrifice ratio ≈ 3.5. Volcker-vs.-Greenspan counterfactual sacrifice ratio = 7.6. The gap reflects ~⅓ of actual disinflation being attributable to supply-shock reversal (oil price fall, dollar appreciation) rather than tight money.
Time-varying NAIRU (Non-Accelerating Inflation Rate of Unemployment): Stable at 5.6–6.3% during 1962–88; falls to 4.5% minimum in 1998; rises to 4.85% by 2004Q4. Omitting supply shocks from the model forces the NAIRU above 8% in the 1970s to absorb otherwise unexplained inflation.
The four-equation "triangle model":
where = inflation rate, = nominal Federal funds rate, = log output gap, = TV-NAIRU (random walk with variance ), = supply shock vector, = real rate target (3%), = inflation target (2%). Sum of lag coefficients on constrained to unity (natural rate hypothesis). Supply shocks include: (i) change in relative price of non-food non-oil imports; (ii) food-energy Personal Consumption Expenditures (PCE) deflator effect; (iii) medical-care PCE effect; (iv) HP-filter productivity trend acceleration; (v) Nixon-era price control dummies. Taylor rule parameters shift at 1979Q2 (Burns→Volcker) and 1990Q2 (Volcker→Greenspan). Serial correlation AR(1) correction via FGLS applied to equation (5).
| Source | Inflation variance pre-1984 | Output gap variance pre-1984 |
|---|---|---|
| Supply shocks | ~80% | ~40% |
| Output error (IS shifts) | ~20% | ~65% |
| Interest rate error | ~0% (eliminated by AR(1) correction) | — |
Taylor rule regimes (AR(1)-corrected):
| Regime | Inflation coefficient | Output gap coefficient |
|---|---|---|
| Burns (1960–79) | ~0.57 | ~0.60 |
| Volcker (1979–90) | ~1.46 | ~0 |
| Greenspan (1990–2004) | ~0.57 | ~0.60 |
"Perhaps the most surprising finding in this paper is that there has been no change in monetary policy after 1990 compared to the policies pursued before 1979, taking a narrow view of policy as the response coefficients in a Taylor Rule monetary policy reaction function."
"Perhaps the most surprising result in this paper is that, when monetary policy is assessed solely in terms of alternative Taylor Rule reaction functions and their effect, there was no difference between the 'Greenspan' monetary policy in effect in 1990–2004 and the 'Burns' reaction coefficients in effect in 1960–79."
Gordon builds a structural story that VAR-based approaches (Stock-Watson SVAR) cannot reach: by naming and measuring specific supply shocks rather than subsuming them in error terms, he can decompose the Great Moderation far more precisely. The serial-correlation finding on the Greenspan reaction function is genuinely striking and under-appreciated — a classic case where an econometric correction overturns a major policy narrative. The main caveat is that supply shock variables (especially import prices) are treated as fully exogenous, but import prices are partly endogenous to U.S. monetary policy through the exchange rate channel. The author acknowledges this, attributing approximately one-third of the oil/dollar supply-shock reversal in 1981–85 to Volcker-era tight money and thus partially re-crediting monetary policy for the disinflation.