Time inconsistency and optimal electoral incentives

January 2, 2013
Many nations have an announced policy of not negotiating? with terrorists?over hostages. Such an announcement is intended to deter terrorists. In other words,? the purpose of the announcement is to influence the expectations of terrorists and thereby their behavior. However, terrorists know that once hostages are taken, policymakers face an overwhelming temptation to make some concession to obtain the hostages’ release. Therefore,?unless the policymakers are credibly?committed?to the policy, the announcement has little effect.? If policymakers couldcommit?to the policy, the incentive for terrorists to take hostages would be largely eliminated. Yet, reneging on the policy when indeed hostages have been taken may be optimal.
Similarly, a father may tell his teenage daughter that he will kick him out his house if she got pregnant out of wedlock. This threat is intended to deter his daughter from getting pregnant. But once?his daughter is?pregnant, it may not be optimal to kick her out of?his house.
The incentive to renege on carrying through a threat when the behavior it was intended to deter? has occurred is refered to?by economists as?time inconsistency or dynamic inconsistency.?A policy may not be consistent across time. What is optimal at time t may not be optimal at t+1.?The concept of time inconsistency may be shed some light on the optimal design of laws or policies.
Prior to the?2012 presidential and parliamentary elections in Ghana, parliament passed a law (C.I.75) that required voters to?go through a biometric verification process before casting their votes. Addressing the media a day before the elections, the chairman of the Eectoral? Commission said “Among the decisions we have taken with the political parties is ‘NVNV- No Verification, No Voting,’ and by verification we mean that everybody would have to be verified biometrically.” This was intended to deter EC officials from allowing people to vote without biometric verification.?If there had been no violation of the law, it would have achieved its deterrent efect. But in reality, some people voted without biometric verification. What should be done??One response is to annul only the votes that were cast without biometric verification. But this is not possible because a person’s vote is not publicly known. Another response is to annul all votes (including the votes of those who were biometrically verified) in affected polling stations. This is the NPP’s preferred course of action in its recent?petition to the Supreme Court.
Followng the NPP’s preferred course of action, the NVNV rule, as a general principle,?means that all votes in affected polling stations should be annuled even if only a very small percentage of them (e.g., 0.01%) was cast by voters who were not biometrically verified; 99.99% of the total votes which were cast through biometric verification should also be annulled.?Is it optimal to?commit to this policy or law at all times??Put differently, is this rule time consistent? Or could it be?time inconsistent in some cases? Perhaps, a commitment to this policy will create the desired deterrent effect in the long run such that the NVNV law will not be violated in future elections. However, as argued below, this is unlikely unless?an obvious?law is enforced.
The deterrent effect of anulling all votes cast at a polling station because an electoral irregularity occured?may paradoxically lead to?a perverse incentive. For example, an EC official who is sympathetic to the NDC may allow votes to be cast without biometric verification in an NPP stronghold in the hope that all votes in that polling station will be annulled; this will give the?NDC an advantage.?Therefore,? a?commitment?to the rule and annulment of votes when the rule is violated lead to?two problems: (i) a selective violation of the rule, and (ii) annulment of valid votes even if a very small proportion of total votes were cast without biometric verification.
The selective violation of electoral laws or rules?in the preceding paragraph should not be dismissed as?a only theoretical possibility or a hypothetical case. It is real. During the recent elections, the following story was posted at the website of peacefmonline:
“A deputy electoral officer and a female technician received severe?beating?from voters at the Ayigya-Ahenbronoum polling centre in the Oforikrom Constituency?over suspicion that they were trying to influence the election results in favour of a particular political party.

The swift intervention of the police patrol team saved them from being lynched.

The victims, Mr Arthur, and Nana Yaa, were accused of deliberately refusing to put the Electoral Commission (EC) stamp on ballot papers given out to voters with the view to invaliding those ballots.

Deputy Superintendent of Police (DSP) Nicholas Opare-Ofosuhene, the Zongo Divisional Commander, confirmed the incident to the Ghana News Agency (GNA), and said the matter was being investigated.”:http://www.ghanabusinessnews.com/2012/12/07/ghana-election-2012-two-electoral-officials-beaten-in-kumasi/

Perhaps, the EC officials were falsely accused. Notwithstanding this possibility, an important implication of the above analysis is that a?commitment?to the annulment of votes in the event of electoral irregularities is not likely to deter the violation of electoral rules and will lead to undesirable?outcomes (e.g., annulment of valid votes even if a very small proportion of total votes was cast without biometric verification). The courts should not be in the business of annuling valid votes or punishing thousands or millions of legitimate voters. This is not politically sustainable.Instead, the state should prosecute?electoral officials who violate the law; in this regard, the court’s business is to?directly and severly punish EC?officials and others who violate electoral laws.?This is the kind?of commitment we need.?It will significantly reduce or eliminate electoral irregularities and save the courts the unpleasant and arguably undemocratic task of annuling thousands of valid votes.?The vigilante justice? that reportedly occured at the Ayigya-Ahenbronoum polling centre exemplifies a weakness in the enforcement of very basic electoral laws.?Relatedly, the enforcement of our electoral laws must be done in a manner that is fair to voters.?For example,?if we want voters to be biometrically verified, let’s make sure the biometric machines are reliable.
Let me conclude by noting that?although this piece was inspired by the NPP’s petition to the Supreme Court,?my ultimate goal?is to help us think about how to improve upon the quality of elections in 2016 and beyond. In a recent article,?H. Kwasi Prempeh opined that??For me,?? this business of an election being challenged in court is about one thing and one thing only: Accountability. Specifically, it is about making sure that the EC understands that it is a body subject to law, and that in carrying out the duties entrusted to it by law?its officials are legally accountable for their acts and omissions. To do less is to allow the EC and its officials to place themselves above the law and beyond the reach of accountability.” Bold font mine.
I couldn’t have said it better. Happy new year to all.
Source: ?J. Atsu Amegashie?


Send your news stories to newsghana101@gmail.com and via WhatsApp on +233 244244807
Follow News Ghana on Google News


Please enter your comment!
Please enter your name here