Should appellate court judges be allowed to delegate the decision making to lower courts?
Why do we let courts delegate the decision-making power by "remanding" to a lower court? Usually remands are justified by arguments such as an efficient division of labor between the court levels (e.g. let the lower court collect evidence and the higher court decide the legal questions). But, some have argued that there are other reasons.
The Law & Economics literature suggests that judges have ideological preferences, which may cause them to deviate from the policy of their superior court (e.g. the U.S. Supreme Court.) A remand mechanism mitigates this problem: judges do not want to exert extra effort when reconsidering the case twice (after a remand) and therefore should respond by not deviating ex-ante.
But... what if judges could circumvent this effort by rolling the cost onto someone else? This can be easily done: appellate court judges get cases from the Supreme court (when the Supreme court remands the case back) but instead of issuing a final decision, they delegate downward an by remanding (again) to the lower (district) court.
A double-edged sword?
In my recent working paper (co-authored with Melanie Demirtas), we argue that granting mid-level appellate courts with the power to remand may be a double-edged sword: one on hand, it helps with the deterrence of the lower court judges (the district courts), because they fear remands and will therefore not deviate ex-ante. On the other hand, it reduces the deterrence of the mid-level court judges (the appellate courts), because those judges can always just delegate downwards instead of exerting the additional effort.
The role of other judicial concerns
We assume that judges have a range of concerns, including ideology, reputation and effort cost. Sometimes these go hand in hand, but sometimes they provide countervailing incentives. We explore how these concerns are translated into the dilemma of whether to remand and test our predictions by empirically analyzing case-level data in the U.S. federal appellate courts.
What are the results?
We find evidence of a "Subsequent Remand Effect" - where mid-level appellate courts are indeed more likely to remand cases that were previously remanded to them by the Supreme Court. We then find that the size of this effect depends on ideological distances between the different court levels, which implies that the effect is at least partially driven by moral hazard.
Check out the working paper
At the following link.