Reports of Death in Puerto Rico are Wildly Exaggerated

A study by a Harvard-led team, published in The New England Journal of Medicine, claims that mortality in Puerto Rico following Hurricane Maria has been far higher than previously thought.  Their study, Mortality in Puerto Rico after Hurricane Maria (the ‘MPR Study’) claims 4,645 excess deaths occurred in on the island from September 20 through December 31, 2017.

How plausible is this?  Not even remotely.

Deaths on the island are reported by the Demographic Registry of Puerto Rico.  Recorded deaths from January to August 2017, the month prior to the hurricane, were essentially flat year on year.  Without additional information, we would expect the count to be flat for the remainder of the 2017 as well.  Notwithstanding, counts were elevated from the September hurricane through November. Based on the latest registry numbers available to us, in September, these excess deaths numbered 540 more than the previous year; October was up 650.  November deaths, however, were up only 113 on the previous year.  In December, the difference was actually negative, a decrease of 649 compared to 2016.   All in, deaths reported by the registry were up 654 for the period covered by the MPR Study.

What caused these excess deaths?  Those persons who were included in the elevated count did not die, for the most part, as a direct result of the hurricane.  That is, they did not drown, were hit by debris or crushed in a building collapse.  That count was reported as 64, and although it could be light, is probably directionally correct.  Rather, excess deaths came from a vulnerable, and principally elderly, population stressed chiefly by a lack of electricity, leading to an inability to access dialysis, respirators and air conditioners.  In addition, some people were unable to reach medical facilities, which may have also been closed for a lack of power.  Finally, some probably succumbed to issues associated with compromised food or water supplies, again related to a loss of power.  In other words, those who were already ill and close to death were pushed over the edge, mostly due to a sustained power outage. 

During any given year, deaths tend to be elevated in December and January.  December 2017 was the exception, almost certainly because many of those who would have died then had died a few months earlier due to the effects of Hurricane Maria.  This explains the surge in deaths in the autumn and the reduction in deaths in December.  Hurricane Maria did not so much kill people outright as accelerate their deaths. 

How can we reconcile 654 excess registry deaths with MPR Study estimates of 4,645 through year-end?

Source:  For Official Deaths, Puerto Rico Institute of Statistics via  Latino USA ; for MPR excess deaths,  Mortality in Puerto Rico after Hurricane Maria , with Princeton Policy monthly interpolations to MPR year-end total.


For Official Deaths, Puerto Rico Institute of Statistics via Latino USA; for MPR excess deaths, Mortality in Puerto Rico after Hurricane Maria, with Princeton Policy monthly interpolations to MPR year-end total.

Perhaps the registry failed administratively to record deaths.  They may have been overwhelmed by work after the hurricane.  Alas, this seems unlikely, because we can see that registry ex-post adjustments are typically 30 or fewer deaths, that is, only about 1% from the original value.  Therefore, recording and reporting are unlikely to be the principal cause of the discrepancy.

Alternatively, perhaps some of the elderly left and died in the US.  The Study’s own numbers suggest this is not the case.  Many young people left the island.  The elderly stayed at home.

We are left with the possibility that the Puerto Rican authorities may have simply overlooked 4,000 bodies.  This is quite a large number to miss.

Imagine, for example, an earthquake hit Los Angeles, a metro area four times the population of Puerto Rico.  Now imagine that many died in the event, with 4,000 people unaccounted for.  What would we expect to see?  First, we would expect most of them to be reported as missing.  As of mid-December, though, only 45 people were still reported as missing in Puerto Rico in the wake of Hurricane Maria.

We would expect to see massive efforts, by both the public and private sector, to locate and identify the bodies.  And we would expect to see on-going news stories about the missing and recovered bodies.  But we see none of that in Puerto Rico.

Thus, we are left with almost 4,000 unaccounted-for persons not recorded as missing, prompting virtually no efforts at recovery of remains, and with no coverage in the media.  This must be considered as highly unusual -- and improbable.

Therefore, unless the Study authors can point to where these bodies may literally be buried, we are left to conclude that they simply do not exist, and the Study must be judged as wildly inaccurate and a gross exaggeration of the true impact of Hurricane Maria in Puerto Rico. 

Instead, total excess deaths may well continue to decline when January numbers come in.  At the year horizon, excess deaths seem likely to settle in the 200-400 range.  This is still a substantial number and a tragedy in its own right.  Most of the fatalities can be attributed to the sustained loss of power experienced by the island. 

This loss was largely of Puerto Rico’s own doing.  The island’s power system is state-owned, and state-owned utilities face political pressures to keep electricity prices low, so low in fact, that they fail to cover needed capital maintenance.  Puerto Rico held its base rate — the part of the electric bill meant to cover operating costs and capital maintenance — unchanged from 1989 to January 2017—nearly thirty years!  Over time, neglected infrastructure becomes vulnerable to unplanned outages.  When incidents occur – as during Hurricane Maria – power generation revenues are inadequate to pay for capital repairs.  That’s why the utility signed a suspect and ultimately stillborn contract with the tiny firm of Whitefish to repair the island’s infrastructure.  That is all the money and credit the utility could muster. 

No matter who owned Puerto Rico’s power utility, given the island’s geography and the massive hit it sustained from Hurricane Maria, power would have been lost for a while.  That the damage was so extensive and repair so delayed is directly attributable to the state ownership of Puerto Rico’s power system and the unrealistically low electricity rates which under-pinned local political popularity not for years, but for decades.

To blame the Trump administration for 4,645 excess deaths is flat out wrong.  Unless the MPR Study authors can actually locate the bodies, 4,000 of those deaths never occurred.  Those that did were largely attributable to the protracted loss of power on the island, unavoidable in part, but made all the worse by the politicized state ownership of Puerto Rico’s power company.