Does The Fed Cause Christmas - Correlation Does Not Equal Causality - The Real Business Cycle Explanation

Sep 21, 2026

Uploaded by Riley

Economic commentary analyzes why many economists, policymakers, and the public mistakenly assume that the Federal Reserve can predictably change real output by adjusting the short-term money supply. It argues that the observed positive correlation between money and output does not prove causality, since temporal ordering alone cannot determine direction. The analysis challenges early macro models and considers alternative explanations, including the possibility that output changes and expectations drive money changes, and questions policy conducted as if the Fed fully controls real output.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages5

Reading Time13 min

AccessFree to download

Does The Fed Cause Christmas?

■ Correlation Does Not Equal Causality

■ The Real Business CycleExplanation

by Charles T.Carlstrom andEdward N.Gamber Economists,policymakers,and eventhe general public frequently make themistaken assumption that the FederalReserve System can effect predictablechanges in real output simply by manip-ulating the short-term money supply.While it is certainly true that a positivecorrelation exists between nominalmoney and real output(see figure 1),thejury is still out on whether this correla-tion indicates that changes in moneycause changes in output. Specifically,it is unproven whethershort-run increases in money abovetrend cause increases in real outputabove its trend level.And perhaps mostimportant,even if money does affectoutput in the short term,it is not clearhow this relationship could be used bythe Federal Reserve to enhance socialwelfare. This Economic Commentary will ex-plore two reasons why the perceptionthat money causes output has main-tained its grip.First,as a general rule,correlation is frequently taken to implycausality—often an incorrect assump-tion.As figure 1 demonstrates,changesin money do precede changes in out-put,inevitably causing some to con-clude a causal relationship.However,simple temporal ordering does not con-cretely indicate causality.Even thoughchanges in nominal money precedechanges in output,that observation inand of itself is not sufficient evidence towarrant any particular conclusion aboutthe direction of causality between # Does The Fed Cause Christmas? money and output.Second,a number ofearly macroeconomic models were con-structed using the assumption that short-term deviations in money would causechanges in output,a view that coincidedwith the conventional wisdom of econo-mists and policymakers.We will offeran alternative explanation for the posi-tive correlation between money and out-put.It argues that although we cannotreject the notion that short-run fluctua-tions in money do not result in short-term fluctuations in output,we also can-not reject the opposite hypothesis thatchanges in real output,or the expecta-tions of future output,cause changes inmoney. After exploring the reasons for thismisperception about money and out-put,this Commentary also looks at thepolicy consequences resulting from it.In particular,since it cannot be knownwith certainty whether the FederalReserve has control over real output,the wisdom of conducting policy as ifit does must be questioned.We con-clude by suggesting an alternativepolicy based on rules. ## ■ Correlation Does Not EqualCausality Does the Federal Reserve cause Christ-mas?A silly concept,perhaps,but con-sider the evidence. As figure 2 illustrates,fourth-quarter changes in the money supply are posi- tively correlated with fourth-quarter changes in output.Without further in- The casual observer,noting thatmoney and output are frequently cor-related,might assume that there is acausal relationship between the two.A closer look at the evidence suggestsotherwise. formation,it is not possible to tellwhether increases in money cause thecorresponding increases in output,orwhether increases in output cause thecorresponding increase in money,orwhether some third,unspecified vari-able causesincreases in both.Knowingonly the positive correlation betweenthe two,an uninformed observer mighteasily conclude that the Federal Re-serve System indeed causes Christmas! The direction of causality is fairly easyto identify in this particular case.Out-put rises in the fourth quarter becauseof increased consumer spending,andthe Federal Reserve boosts the moneysupply to accommodate the increasedspending.However,this does not com-pletely explain the causal relationshipbetween money and output,becausethe two are correlated at times otherthan Christmas as well. And even though figure I clearly indi-cates that changes in money precedechanges in output,can we concludefrom this that changes in money"cause"changes in output?Unfortu-nately,inferences about the direction of causality between two variables cannotbe based on thei

FAQs

Why is the correlation between money and output not enough to prove causality?

Because correlation can be mistaken for causality, and temporal precedence alone cannot establish which variable causes the other or whether a third factor is responsible.

What problem arises when trying to infer causality from temporal ordering?

Temporal ordering may reflect reactions to other information; for example, umbrella-carrying and rain can be driven by forecast awareness rather than one causing the other.

What policy implication does the commentary draw from uncertainty about Fed control over real output?

Since it cannot be known with certainty that the Federal Reserve controls real output, conducting policy as if it does should be questioned, and an alternative rule-based policy is suggested.

Sep 21, 2026

Uploaded by Riley