Fed's Money Supply Policy: A Regime Shift

The core objective of this research is to analyze whether the noticeable shift in U.S. Federal Reserve monetary policy between the pre-1979 (pre-Volcker) and post-1979 (Volcker-Greenspan) periods is reflected in adjustments to the high-powered money supply. The study focuses on nonborrowed reserves, a key monetary instrument, and examines how these reserves respond to macroeconomic indicators. The analysis employs both ex-post and real-time data to provide a comprehensive view of the Federal Reserve's actions. Real-time data, including Greenbook forecasts, offer insights into the information available to policymakers at the time of decisions. This approach aims to uncover systematic changes in nonborrowed reserves in response to changes in expected inflation and the output gap. The study then investigates if these observed patterns align with a welfare-maximizing policy framework and whether they ensure equilibrium determinacy, which is a key theoretical aspect explored in relation to the observed policy responses.

The study directly compares the Federal Reserve's monetary policy actions in the pre- and post-Volcker periods. The analysis shows a marked difference in how the money supply responded to changes in inflation. In the pre-Volcker era (before 1979), the inflation feedback on the money supply was positive, indicating a less aggressive anti-inflationary stance. This contrasts with the post-Volcker period (after 1979), where the inflation feedback was negative. This shift strongly suggests a change in the Federal Reserve's approach to inflation management. Notably, the response to the output gap remained consistently negative across both periods. The consistency of this negative relationship across both periods highlights the Federal Reserve's focus on output stabilization, regardless of the prevailing policy regime. The analysis further investigates the robustness of these findings by utilizing various specifications of the forward-looking component (inflation) and the output gap, along with real-time data to ascertain if the observed shifts are consistent across different modeling approaches.

The paper utilizes data from the Federal Reserve Bank of St. Louis, covering a forty-year period (1960:1-1999:4) at a quarterly frequency. Inflation is measured using both the GDP deflator and the Consumer Price Index (CPI). The analysis estimates forward-looking reaction functions for nonborrowed reserves, mirroring similar approaches used to study the federal funds rate. The paper explicitly addresses criticisms related to using ex-post data by incorporating real-time data, specifically the Greenbook forecasts prepared by the Federal Reserve staff for the Federal Open Market Committee (FOMC) meetings. The use of Greenbook forecasts provides a more precise picture of monetary policy decisions, as it reflects the information available to policymakers at the time. The analysis also employs vector autoregressions (VARs) and the Generalized Method of Moments (GMM) technique for estimation, reflecting standard econometric methods for examining the links between monetary policy instruments and macroeconomic variables.

The study uses a standard macroeconomic model with sticky prices to analyze the equilibrium determinacy properties of different monetary policy regimes. A crucial finding is that both the pre- and post-Volcker policy approaches lead to equilibrium uniqueness, implying that neither regime allowed for endogenous fluctuations driven by non-fundamental shocks. This conclusion stands in contrast to earlier research suggesting that the pre-Volcker policy might have been associated with multiple equilibria, potentially contributing to higher inflation volatility. The model reveals that the key requirement for achieving macroeconomic stability is a money supply policy that ensures a decrease in real balances as inflation increases. This condition, coupled with a negative response to the output gap, underpins the determinacy of both regimes and demonstrates the theoretical consistency of the Federal Reserve's actions, even amidst the apparent shift in approach observed during the Volcker-Greenspan era.

The core of the empirical analysis involves estimating forward-looking reaction functions for the growth rate of nonborrowed reserves. This approach, similar to methods used in studies of federal funds rates, directly examines the systematic relationship between monetary policy and macroeconomic indicators. The study uses a single-equation approach, a parsimonious specification in growth rates, which provides a clearer interpretation of the estimated coefficients. Unlike previous vector autoregression (VAR) studies that indirectly considered the impact of future inflation, this analysis models expected inflation explicitly, along with the output gap, to provide a more precise estimation of the monetary policy reaction function. The choice to use nonborrowed reserves, rather than broader aggregates, allows for a focused examination of the Federal Reserve's direct control over this key monetary instrument. The study uses data from the Federal Reserve Bank of St. Louis, spanning 1960:1-1999:4, with the benchmark inflation measure being the annualized percentage change in the GDP deflator; however, alternative measures, including the Consumer Price Index (CPI), are also considered.

The empirical results reveal significant differences in the Federal Reserve's money supply adjustments between the pre-1979 (pre-Volcker) and post-1979 (Volcker-Greenspan) periods. Consistent across both periods is a negative response of nonborrowed reserves to a widening output gap. However, a crucial difference lies in the response to inflation. In the pre-Volcker era, the estimated coefficient on expected inflation is positive, although not statistically significant, indicating a less pronounced anti-inflationary monetary policy response. This contrasts sharply with the post-Volcker era, where the response to expected inflation is significantly negative, suggesting a strong commitment to inflation stabilization. This finding confirms previous evidence based on federal funds rate analyses indicating a clear shift in the Federal Reserve's policy stance, reflecting a more active management of inflation. The study verifies that these findings are robust across alternative specifications of expected inflation, output gap measures, and different inflation target horizons.

To enhance the robustness of the findings, and address concerns about biases from using ex-post data, the analysis incorporates real-time data. Specifically, the study utilizes Greenbook forecasts, prepared by the Federal Reserve staff for FOMC meetings. These forecasts offer several advantages: they represent the information available to policymakers at the time of decisions, potentially have an informational edge over private sector forecasts, and are less susceptible to the Lucas critique by accounting for potential structural changes in the economy. The results using these real-time data confirm the main findings from the ex-post data analysis: a significant difference in the response to expected inflation between the two periods, with a positive response in the pre-Volcker era and a negative response in the Volcker-Greenspan era. The consistency of results between ex-post and real-time data underscores the reliability of the findings and suggests that the observed regime shift is not an artifact of data revisions.

The theoretical analysis employs a standard sticky-price macroeconomic model to examine the implications of the estimated money supply reaction functions for equilibrium determinacy. The model incorporates features such as endogenous labor supply, an additively separable CES utility function, and staggered price setting (à la Calvo, 1983). Money demand is introduced through real balances entering the utility function, and uncertainty is represented by cost-push shocks. The analysis focuses on local equilibria that converge to a long-run equilibrium. The equilibrium conditions are log-linearized around the steady state, assuming a sufficiently small support for aggregate shocks and a steady-state gross nominal interest rate greater than one. This simplification allows for a tractable analysis of the dynamic properties of the model under different monetary policy rules, specifically focusing on the relationship between the money supply and inflation.

The model incorporates a forward-looking money supply reaction function, where the money growth rate responds to expected future inflation and the output gap. This setup is analogous to the forward-looking interest rate rules found in existing literature. The analysis investigates the conditions under which this money supply reaction function is consistent with the central bank's optimal commitment plan under a timeless perspective. The model suggests that for optimal policy, nominal balances should respond to expected inflation by less than one-to-one ( < 1), and should decrease with the output gap (y < 0). This contrasts with the so-called 'Taylor principle' commonly associated with interest rate rules. The absence of a lagged instrument in the money supply reaction function is justified by the inherent backward-looking element already present in the model (m^b_t-1), which suffices to implement history-dependent equilibrium sequences.

The analysis focuses on the conditions necessary for equilibrium uniqueness in the model. Equilibrium uniqueness requires the existence of exactly one stable eigenvalue, which imposes restrictions on the coefficients of the money supply reaction function. The paper derives these conditions for both current-looking (n=0) and forward-looking (n=1) reaction functions. Applying these conditions to the empirical estimates, the study shows that both the pre-Volcker and Volcker-Greenspan monetary policy regimes ensured equilibrium uniqueness. This is demonstrated by showing that the estimated feedback coefficients satisfy the conditions for a unique rational expectations equilibrium, regardless of whether the data used are ex-post revisions or real-time observations. This finding contradicts previous studies emphasizing the potential for indeterminacy under pre-Volcker policy, demonstrating that a money supply approach provides a different perspective on policy stability.

The study's empirical analysis reveals a clear regime shift in US Federal Reserve monetary policy between the pre-1979 (pre-Volcker) and post-1979 (Volcker-Greenspan) periods. The key finding is a significant difference in how the money supply (nonborrowed reserves) responded to changes in expected inflation. In the pre-Volcker era, the response was positive, suggesting a less aggressive, potentially more accommodating, approach to inflation. Conversely, in the post-Volcker era, the response was significantly negative, indicating a stronger commitment to inflation stabilization. Notably, the response to the output gap remained consistently negative across both periods, revealing a persistent focus on output stabilization regardless of the inflation-targeting approach. This conclusion remains robust to changes in model specification and the use of real-time data, specifically Greenbook forecasts, strengthening the confidence in the findings of a clear policy shift.

The theoretical framework, a standard sticky-price model, explores the implications of the empirically observed money supply reaction functions for equilibrium determinacy. The model reveals that a crucial condition for macroeconomic stability is that real balances decrease as inflation increases. Importantly, this condition is satisfied under both the pre- and post-Volcker policy regimes. The theoretical analysis definitively concludes that despite the different responses to inflation, both periods maintained a uniquely determined rational expectations equilibrium. This finding directly contradicts some earlier research that suggested potential indeterminacy (multiple equilibria) under pre-Volcker policy. The study demonstrates that, viewed through the lens of money supply, the Federal Reserve's policy actions, even before Volcker's chairmanship, did not result in macroeconomic fluctuations driven by non-fundamental shocks.

The study successfully integrates its empirical and theoretical findings to offer a comprehensive assessment of US monetary policy. The empirical evidence shows a clear shift in the Federal Reserve’s response to inflation, transitioning from a more accommodating stance in the pre-Volcker era to a highly anti-inflationary one in the Volcker-Greenspan era. The theoretical analysis, however, reveals that both regimes ensured equilibrium determinacy, challenging earlier conclusions drawn from interest rate rule analyses. The study ultimately finds that the pre-Volcker policy, while less explicitly anti-inflationary based on the money supply reaction function, did not contribute to macroeconomic instability due to multiple equilibria. This highlights the importance of considering the money supply mechanism, specifically the role of nonborrowed reserves, when analyzing and interpreting the effectiveness and stability of Federal Reserve monetary policy.