Nuclear disasters and black money amnesty: How probability fools us
The Japanese Nuclear Commission had the following goals set in 2003: “The mean value of acute fatality risk by radiation exposure resultant from an accident of a nuclear installation to individuals of the public, who live in the vicinity of the site boundary of the nuclear installation, should not exceed the probability of about 1x10⁻⁶ per year (i.e. at least 1 per million years)”.
That policy was designed only 14 years ago. The one in a million-year accident almost occurred about 8 years later when the Fukushima I Nuclear power plant underwent a nuclear meltdown.
In his presidential commission report on the NASA Challenger disaster, the theoretical physicist Richard Feynman was struck by management’s claim that the risk of catastrophic malfunction on the shuttle was 1 in 100,000, which would entail that NASA could expect to launch a shuttle every day for the next 274 years while suffering, on average, only one accident.
The Challenger space shuttle orbiter broke apart 73 seconds into its flight.
So what does a nuclear and a space disaster have to do with black money and tax amnesties?
The answer lies in our struggle to understand probabilities of events, and more so when we are dealing with small probabilities associated with the occurrence of rare events. Mankind’s experience in the area of nuclear and space events is merely eight decades. Simply put, we just do not have enough experience in these fields to accurately estimate the probabilities of aforementioned disasters.
Tax amnesties are meant to provide a window to dishonest taxpayers to come clean. Like nuclear and space disasters, tax amnesties are supposed to be rare events. It should be rare for an individual, in the context of an average taxpaying period of about forty years. So ideally, an amnesty should be available to a taxpayer only once in his lifetime, thereby making it a rare event for him. This rarity is chiefly responsible for miscalculation of probabilities. While the miscalculation works negatively in the space of nuclear and space projects, it works positively in case of tax amnesties. As prospect theory informs us, people comply to amnesty schemes because they overestimate the likelihood of being caught, essentially because we are poor at calculating probabilities of a rare event. Small probabilities are over weighted in terms of their impact on decisions which makes amnesties a useful tool. However, even though progressive amnesty schemes have been punitive, we have ignored one important aspect of these schemes: the frequency with which they have been made available. In the last three years, this is the fourth such scheme, with two schemes in 2016 (IDS ending September 2016 and the current scheme post demonetization — both targeting domestic black money), one in 2015 (foreign black money) and one in 2013 (service tax).
The regularity of an amnesty scheme changes this very probability. The increasing frequency reduces the rarity of the event, thereby making us better at estimating the probability. With a one off scheme, the decision to declare black money is taken under ambiguity (uncertainty with unknown probabilities). As they become more regular, the decisions are subsequently taken under risk (uncertainty with known probabilities). We move decision making from the domain of ambiguity to the domain of risk, essentially signaling that the next scheme is around the corner.
In effect, as the probability of an amnesty rises, and thus the future opportunity to declare any dishonesty free of penalty, people report less income, thereby actually decreasing, rather than increasing, the efficiency and equity of the tax system. By constantly introducing amnesty schemes, instead of mopping up tax on hitherto undeclared income, the government may just be collecting tax on income which would have been reported in time, but is being reported only now. In a survey on the attitudes of taxpayers towards tax compliance and tax amnesties in South Africa, the findings indicated that the offering of multiple tax amnesties might not generate additional revenue, as non-compliant taxpayers will continue to evade taxation in anticipation of additional future amnesties. Experimental studies within taxation psychology have reached strong findings about the fact that tax amnesties have negative effects on tax compliance.
The other worrying aspect is that subsequent amnesty schemes see participation from the same set of people. In India, it has been determined that taxpayers, who apply for a tax amnesty, are generally the same people. Similarly, it was witnessed that the same people and groups (institutions) generally get benefit from tax amnesties which have been promulgated 21 times in a 35-year period in Argentina.
Hence, instead of bringing ‘newer’ people under the tax purview, we are merely providing the existing black money hoarders a free run. Additionally, the negative psychology that tax amnesty creates in justice perceptions of honest taxpayers may interrupt voluntary compliance process in the future (long-term). Because there is no guarantee that new amnesty practices are not promulgated by the governments and there may also be a considerable amount of taxpayers who have made a habit of tax amnesty expectation. While increased tax compliance is desirable, the constant use of amnesty schemes may not be the right tool to drive it.
