For the last twenty-two years, I have written well over 25 opinion essays in different newspapers or online news media. I have also experimented with blogging. In January 2016, I started Politics and Snapshots, a blog on my two hobbies: photography and American politics. It was a fun experiment! But given my teaching and administrative duties, it was difficult to produce content on a regular basis.

In October 2017, a month after Hurricane Maria devastated Puerto Rico’s infrastructure, I started a new blog, The Puerto Rico Data Lab. This experiment was different from the previous one in two important respects. First, I launched the blog after deciding to research and write on U.S. – Puerto Rico relations in the aftermath of Hurricane Maria. Thus, the blog served as a personal notebook where I could sketch ideas and share them with the world as I prepared articles for academic journals. Some of these posts were also published in different online news outlets. Second, I used the blog to further develop my data analytic skills, learning R and other data visualization packages.

On this site, I will share short commentaries of my research interests and my reactions to different political events.

Getting Ready for the 2020 Democratic Presidential Nomination Caucuses and Primaries

Monday, February 3, 2020, Iowa will hold their caucuses, marking the official start of the Democratic presidential nomination contest. New Hampshire’s primary will be held eight days later, followed by 55 more contests ending on June 6, 2020. 

Although 12 candidates are competing for the Democratic Party’s presidential nomination, it seems that three have a realistic shot at winning the nomination: Joe Biden, Bernie Sanders, and Elizabeth Warren.

Recent polls show that Biden is still in the lead, but Sanders is closing the gap, while Warren’s numbers keep sliding. These polls also indicate that Mike Bloomberg’s popularity is increasing and that Andrew Yang’s is gaining more support. Could Pete Buttigieg or Amy Klobuchar pull off a surprise win in one of the first contests?

Although they represent 4% of all pledged delegates, the first four contests earn lots of media attention. For the weaker candidates, a poor showing in these primaries or caucuses will end their presidential hopes while the winners will get a bump in the polls. But for the top candidates, the real test will be on March 3, Super Tuesday. On this day, 15 states, including California and Texas, will hold their primaries and the candidates will compete for 34% of all the pledged delegates.

Other dates in the schedule are important too. On March 17, the candidates will compete for 15% of the pledged delegates in four major major primaries. If a winner has not been decided by then, the Acela Primaries on April 28 will likely determine the presumptive nominee. 

Who will win the nomination? It is too early to say. But the answer will start to be clearer after Super Tuesday. And for those who are curious about upcoming contests, here is a list of this year’s Democratic caucuses and primaries.

Increasing Puerto Rico’s Electricity Rate Will Negatively Affect its Economic Recovery in the Short-Term

One of Puerto Rico’s main economic challenges is the high cost of electricity. As I noted in an early post, the island’s residents pay one of the highest electricity rates in the United States. For the past three years, the Financial Oversight and Management Board (FOMB), established by the U.S. Congress to oversee Puerto Rico’s finances and to restructure its debt, have instructed the Puerto Rico Electric Power Authority (PREPA) to set the the price to 21 cents per kilowatt/hour (kWh) by 2023. Ricardo Rossello, Puerto Rico’s pro-statehood governor who wants to privatize PREPA, promised last year that PREPA would to reduce the rate to 20 cents kWh. The Puerto Rico Manufacturers Association in an effort to increase its members’ competitiveness have lobbied the island’s government to reduce the electricity rate to 15 cents kWh.

Can PREPA deliver lower electricity rates? In the short-term, it will be unable to meet Rossello’s 20 cents kWh target. If a new deal to restructure a part of PREPA’s $9 billion is approved by Judge Laura Taylor Swain and Puerto Rico’s legislature, PREPA will pay off the new bonds by levying its customers a new surcharge. Starting this summer, the electricity rate will increase by 1 cent to 23 cents per kWh. In July 2020, the surcharge increases to 2.8 cents per kWh and over the years it will keep increasing up to 4.6 cents per kWh until the bonds are paid off.

Although many Puerto Ricans have voiced their opposition to this deal, Governor Rossello expects that the projected hikes will be balanced by his government’s reform of PREPA, which includes the privatization of its assets and the conversion of some of its power plants from oil to natural gas. While Puerto Ricans may see lower electricity rates in the future, it is clear that the price for electricity will increase in the short-term.

These rate hikes will be an extra burden on Puerto Rican families and businesses. While Puerto Rico’s residents do not pay the highest electricity rate in the United States, the island’s median household income is really low. Meaning that Puerto Ricans tend to devote higher percentage of their income to pay for electricity.

The next graph compares Puerto Rico’s median household income for 2017 with the median household income for the U.S. It is worth noting that these are the latest figures, calculated by the U.S. Census.

To put it perspective, the median American family spent 2.4% of their household income on electricity. At 29.5 cents per kWh, Hawaii’s price for residential electricity in 2017 was the highest in the United States. But, the state’s median household income was $74,493. Thus, Hawaiians were spending around 2.4% of their income on electricity.

For example, the residential electricity rate in Hawaii for 2017 was 29.5 cents per kWh. But the state’s median household income was $74,493. Thus, Hawaiians were spending 2.4% of their income on electricity.

Using this formula, we notice that Puerto Rico’s median family spends 4.2% of their income on electricity.

YearMedian Household IncomePercentage of Median Household Income Devoted to Electricity

While the amount of income devoted to electricity is lower than in 2015, the new rates hikes will reverse recent gains. The next graph already includes the proposed increase for 2019. Thus, Puerto Ricans will once again experience a steep rise on electricity costs.

While forecasting the future impact these rate hikes will have on the Puerto Rican economy is difficult, the sad reality is that at least in the short-term the rising cost of electricity will burden the island’s economy, squeezing family incomes, forcing some businesses to close, or hampering the Puerto Rican government’s efforts to attract new investments.

Note: A version of this post can be found in my blog, The Puerto Rico Data Lab. Follow me in Twitter: @cyordan or @pr_datalab.

Comparing Average Electricity Prices in Puerto Rico and the United States, 2009-2018

I recently returned from Puerto Rico, where I was presenting a paper connected to my research in an academic conference on the island’s recovery following Hurricane Maria. I also took the opportunity to see the family.

Although things are improving since my last visit in August 2018, Puerto Rico still faces many economic problems, mostly connected with the debt crisis that preceded the 2017 hurricane season. A few days ago, the Financial Oversight and Management Board (FOMB), established by the U.S. Congress in 2016 to stabilize Puerto Rico’s finances, and Governor Ricardo Rosselló announced a new deal to restructure around $3 billion of the Puerto Rico Electric Power Authority’s (PREPA) $9 billion debt.

If this deal is approved by Judge Laura Taylor Swain and Puerto Rico’s legislature, PREPA will pay off the new bonds by levying its customers a new surcharge. Starting this summer, the electricity rate will increase by 1 cent per kilowatt hour (kWh) to 23 cents per kWh. In July 2020, the surcharge increases to 2.8 cents per kWh and over the year it will keep increasing up to 4.6 cents per kWh until the bonds are paid off.

While the deal helps PREPA reduce its debt obligations, freeing up cash to pay for other costs or to make future investments to the electrical system, the foreseen rate increases will hurt Puerto Rican consumers. Rising energy costs will also negatively affect the island’s businesses and potentially discourage future investments.

This deal got me thinking about the average price for electricity for PREPA’s residential, commercial and industrial consumers and how it compares with average rates for these consumers in the United States. The following graphs summarize these rates for the last ten years (2009-2018). All the figures have been adjusted for inflation using the Consumer Price Index for 2018 as the base rate.

What can we learn from these graphs? Here are three quick observations.

  • The average price for the United States is a bit deceiving as many Americans pay higher rates for their electricity. However, it is critical to notice the stability of these prices. In Puerto Rico, fluctuations in electricity prices is an indicator of PREPA’s inability to effectively manage the production, transmission and distribution of electricity.
  • Puerto Rico’s price differentials are problematic. On average, commercial customers in the U.S. territory are paying a lot more for electricity than in the mainland. The same holds true for industrial firms. This reality may explain why many businesses have not expanded their operations or reduced the prices of their goods or services.
  • Future increases in these rates will burden the island’s economy, squeezing family incomes, forcing some businesses to close, or hampering the Puerto Rican government’s efforts to attract new investments.

What do you think of these figures? Do you think that the new deal to restructure part of PREPA’s debt is a wise choice?

The 2016 BREXIT Vote: Part Two

In my last post, I explained the origins of the BREXIT vote, the inability for pollsters to forecast the Leave Campaign’s surprising victory, the regional distribution of the vote and the potential impact BREXIT could have on the United Kingdom’s future territorial unity. In this post, I want to breakdown the BREXIT vote along demographic lines. The goal is to understand who supported BREXIT and who wanted Britain to remain in the European Union (EU).

In the event, the British Parliament considers to hold a second referendum, which at this time is very unlikely, these numbers will help each side of the debate mobilize their supporters.

A Generational Divide

As noted yesterday, 72% of the British electorate participated in the BREXIT referendum. Let us first look at the participation rate by age group.

How did voters in each group vote?

Clearly, younger voters were more supportive of the Remain Campaign than the Leave Campaign. Given higher turnout rates among older voters, they helped the Leave Campaign win the referendum.

This generational divide is a bit daunting for the UK’s future. As it starts to recast its economic future, the risks associated with this move will impact younger people more than older voters. It is worth highlighting that 75% of voters between the ages of 18-24 voted for the UK to remain in the EU. That is a big margin.

Men and Women on BREXIT

Here is a table that summarizes the results of the vote along gender lines. Men participated at higher rate than women and they overwhelmingly voted to leave the EU. Women had mixed views on the EU.

Gender Remain Leave Turnout
Male 45% 55% 74%
Female 51% 49% 71%

Race and the BREXIT Vote

Immigration issues was one of the big drivers of the BREXIT vote. For those voters who felt that the UK should regain control over its border and immigration policies, BREXIT was seen an important step. Those people who favored the EU’s open borders and its labor mobility regulations voted to stay. I will write about this in post in the future.

But the debate on immigration activated identity politics. The next table captures an important racial divide regarding BREXIT.

Race/Ethnicity Remain Leave Turnout
White 46% 54% 74%
Black or Minority Ethnic Group 69% 31% 58%

It is clear that White British voters overwhelmingly supported the Leave Campaign, while minority voters strongly voted for the Remain Campaign. Could a higher turnout rate among minority voters help keep the UK in the EU? The number of minority voters in the UK, compared with the United States, is very small. Thus, a higher participation would not have made a big difference.

The BREXIT Vote and Social Class

To measure voters’ social class status, we can use a voter’s type of employment as a proxy. The British Census includes four key classifications.

Let us look first at the turnout rate for each of these classifications.

While the turnout rate among the highest classes was higher than the national average, the fact of the matter is that these voters represent a smaller size of the electorate. By in large, voters in higher income professions were more apt to support the Remain Campaign than people in lower-paid professions, as the next graph demonstrates.

The Education Divide

In terms of voters’ educational background, the turnout rate for voters with the U.S. equivalent of a high school education or less (i.e. “no qualifications”) was estimated at 70%, while it stood at 71% for voters with some tertiary education. The rate for voters who completed a university degree or higher was 78%. The next table summarizes the distribution of BREXIT votes along these three groups.

The higher voters’ education attainment, the more critical they are about BREXIT.

Political Affiliations and BREXIT

Finally, let us look if voters’ affiliations to Britain’s main, national political parties had any impact on their vote in the referendum.

Even though the UK Independence Party is not a major political party and it has a very low number of follower, when compared to the other parties included in the graph above, it is worth remembering that its leader, Nigel Farange was one of the main leaders of the Leave Campaign. It is impressive that close to 100% of its voters want Britain to exit the EU. Support for Brexit was stronger in the Conservative Party than in the Labour Party and Liberal Democratic Party.

Not surprisingly, the turnout rate for people affiliated to one of the main political parties was higher than the national average.

Concluding Thoughts

As Britain gets closer to its expected departure from the EU, it is worth remembering that many Britons voted in favor of remaining in the EU. Should the British parliament consider a new referendum? For now, the House of Commons has rejected proposals to hold a new referendum. But assuming that the members of parliament had a change of hearts, would voters change their vote? It is difficult to answer these questions. BREXIT divided the public and these divisions will define British politics for some time.

The 2016 BREXIT Vote

I will be giving a talk on Britain’s decision to leave the European Union (EU). After several years of negotiations between the British Government and Brussels, it seems that for now the United Kingdom’s exit will be on 12 April 2019.

How did we get here?

Britain’s relationship with the EU has always been tenuous. Historically speaking, it favored deeper integration of the regional bloc’s common market, but it balked at supporting any policy that would impinge on the United Kingdom’s political, economic or financial sovereignty. Thus, the UK did not join the European Monetary Union, which established the European Central Bank and the euro. It also blocked attempts to strengthen the European Commission’s powers or the European Parliament’s competence in different policy areas.

Over the years, Euro-skeptics in both the Conservative and Labour parties had requested the British Government to renegotiate the UK’s status within the EU. For example, in the 1980s, Margaret Thatcher’s Government negotiated a British reduction to the European Community’s budget, known more commonly as the “UK correction” or the “UK Rebate”.

By 2012, there was a growing number of people in the Conservative and Labour parties, and within the UK’s citizenry, asking for a referendum on Britain’s membership on the EU. While Prime Minister David Cameron agreed that the UK should renegotiate its status within the EU, he opposed holding a national referendum. But, carrying out these negotiations were difficult because Cameron’s Conservatives had formed a coalition Government with the pro-EU Liberal Democratic Party.

In 2015, Britain held parliamentary elections and sensing strong Euro-skeptic sentiments amongst the electorate, Cameron and his Conservative Party promised to hold a referendum on the Britain’s membership in the EU, if his party won a clear majority. Cameron’s decision was in part influenced by the rising influence of the UK Independence Party (UKIP), a fringe right-leaning, anti-EU party.

After winning a majority of the vote, Cameron’s Government started to negotiate new concessions with the EU on multiple issues and it called for a referendum on Britain’s membership in the EU, which was held on 23 June 2016.

Public Opinion Polling and the Results of the Referendum

Public opinion data, as can be seen in the following graph, suggested that the Remain Campaign, led by David Cameron and other British politicians, including Theresa May – the current Prime Minister, would win 52% of the vote. The Leave Campaign, which was most associated with the views of the UKIP’s Nigel Farage and the Conservative Party’s Boris Johnson, was projected to win 48%.

The polls did not capture the Leave Campaign’s energy. It won 52% of the vote, while the Remain Campaign captured the remaining 48%.

While the polls failed to forecast British support for the Leave Campaign, by the end of the campaign the pro-BREXIT had been able to mobilize its supporters. To show this, the next graph uses the first graph’s data, but it includes the trendlines for the two campaigns.

These trendlines clearly show that the race was closer than the final public polling averages suggested.

Today, it is difficult to understand the BREXIT vote’s long-term implications. But breaking down the results of the referendum will help us understand what sort of challenges Britain faces in the near future.

Deconstructing the BREXIT Vote

Overall, 72% of the British electorate cast a vote during the referendum. To put this number in perspective, it is worth noting that the rate of participation was higher than any of the parliamentary elections held since 1997.

Let us look first at how electors in the UK’s territorial units (i.e. nations and provinces) voted.

What was voters’ preferences in these territorial units?

As the bar graph shows the Leave Campaign won a majority of the vote in Wales and England. But it is important to remember that most British citizens live in England. Thus, English voters, more so than voters in the rest of the country, decided Britain’s fate with the EU.

RemainLeaveTotal Votes%
N. Ireland440,707349,442790,1492%

Here is a breakdown of the vote by regions.

The data shows that while voters in London preferred to remain in the EU that was not the case in rest of England.

Concluding Thoughts

What can we learn about the referendum’s results? Differences between Britain’s territorial units, will foster two problems. First, on Britain exits the EU, we can expect Scottish leaders to call for another independence referendum. One the reasons why the 2014 Scottish referendum failed was Scots’ fear that independence would mean that Scotland would not automatically be part of the EU.

While most voters in Northern Ireland wanted to stay in the EU, the vote was cast along sectarian lines. Thus, 85% of the Catholics/Nationalists voted for the Remain Campaign, while 60% of Protestants/Unionists voted for the Leave Campaign. Many people Brits fear that reestablishing a border to separate the Northern Irish counties from Ireland will reignite the sectarian conflict that ended with the signing of the 1998 Good Friday Peace Agreements. This one of the reasons that Prime Minister May’s Government is trying to find a way to leave the EU but without reestablishing a border.

In part two of the analysis, we will breakdown the 2016 Brexit vote along demographic lines. For now, we can conclude that Britain’s future territorial integrity is in question.

Do a Majority of Americans Want to Get Rid of the Electoral College?

In my last post, I considered ways to reform the Electoral College (EC). One way is for states to join the National Popular Vote Interstate Compact (NPVIC). The other option is for states to adopt the Congressional District Method which Maine and Nebraska currently use to apportion their electoral votes.

I noted in my analysis that it was not clear what Americans currently think of the EC or the NPVIC. Luckily, this week’s The Economist/YouGov Poll asked its panel a number of questions connected to this debate. The results tend to be in line with Gallup’s 2016 findings, which I discussed in my last post.

The Economist/YouGov Poll asked three questions.

While the last of these questions did explain how the NPVIC works, it is surprising how many Americans are not sure whether or not they like the idea.

What can we learn from this poll’s findings? Here are three observations, which largely reflect the country’s political mood.

  1. Democrats do not hold too much regard for the EC and they want to amend the Constitution to eliminate the system and let voters directly elect the president.
  2. Republicans have favorable views of the EC and prefer to keep the system as is.
  3. Independents are divided on this issue and many of them are not sure what to think of the current electoral system.

Given these divisions, the EC is here to stay! The best the EC’s critics can hope for is for states to reform how they award their electoral votes. Clearly, the NPVIC is one possible reform, but most people are not aware of this initiative. The other option is for states to adopt the Congressional District Method I reviewed in my last post. A final proposal, which I did not discuss in my proposal and has been discussed by considered by some state legislatures, is for states to apportion their electoral votes proportionally.


Should We Reform the Electoral College? A Few Considerations

For the last days, I have been thinking about the way Americans elect their president and I am not the only way thinking about this issue, according to Google Trends data.

To win the presidency, a nominee has to earn a minimum of 270 electoral votes, rather than a majority of votes cast. This is not the place to review the history of the Electoral College (EC), but it is important to keep in mind that it totals 538 votes. Each state and the District of Columbia gets a proportional share of these electoral votes, according to the following formula:

  1. States receive two electoral votes, which corresponds to each’s allotment of two seats in the U.S. Senate.
  2. The remaining electoral votes are distributed among the 50 states based on each’s number of Congressional Districts.
  3. With the adoption of the 23rd amendment to the U.S. Constitution, the District of Columbia has three electoral votes.

The EC is a much-maligned system. In a recent town-hall, Massachusetts Senator and presidential hopeful Elizabeth Warren explained why she wants to get rid of the EC and let Americans vote directly for their president. Here is one of Senator Warren’s tweets summarizing her position:

Unsurprisingly, Republicans oppose her proposal. For example, South Carolina Senator Lindsey Graham tweeted:

Even President Donald Trump, who won the EC but lost the popular to Hillary Clinton by almost 3 million votes, entered the debate, sharing his thoughts on Twitter.

What do Americans think of the EC? Do they want to get rid of it? Last summer, the Public Religion Research Institute (PRRI) conducted a poll that found that close to two-thirds of Americans want to directly elect their president by popular vote. It also demonstrated that Democrats strongly favored getting rid of the EC while Republicans’ attitudes are mixed. A Rasmussen survey of likely voters carried out in early October 2018 did not corroborate the PRRI’s findings. Only 46% of respondents favored “eliminating” the EC. In December 2016, Gallup found that 49% of Americans wanted “to amend the Constitution to allow for a popular vote for president”, but this was lower than the 62% of respondents who in 2012 wanted to get rid of the EC. In addition, Gallup’s research also suggested that 81% of left-leaning and 19% of right leaning Americans wanted to do away with the EC.

While many Americans are dissatisfied with the EC, amending the constitution is unrealistic in today’s hyperpartisan environment. So what can we do?

One solution, as Jamelle Bouie recently argued in the pages of The New York Times, is for states to join the National Popular Vote Interstate Compact. This initiative is quite simple. It requires states (and the District of Columbia) to pledge their delegates to the winner of the national popular vote. So far, 13 states, equaling 181 electoral votes, have passed legislation supporting this project. But the system goes into effect once the initiative attracts 270 electoral votes.

StateGovernorState Legislature2016 Pres. WinnerVote Margin
CADemocratDemocratic MayorityHillary Clinton30%
CODemocratDemocratic MayorityHillary Clinton5%
CTDemocratDemocratic MayorityHillary Clinton16%
DCDemocratDemocratic MayorityHillary Clinton86%
HIDemocratDemocratic MayorityHillary Clinton32%
ILDemocratDemocratic MayorityHillary Clinton17%
MARepublicanDemocratic MayorityHillary Clinton27%
MDRepublicanDemocratic MayorityHillary Clinton26%
NJDemocratDemocratic MayorityHillary Clinton14%
NYDemocratDemocratic MayorityHillary Clinton25%
RIDemocratDemocratic MayorityHillary Clinton16%
VTRepublicanDemocratic MayorityHillary Clinton26%
WADemocratDemocratic MayorityHillary Clinton16%

While other states are considering joining the Compact, the biggest hurdle, as the table shows, is that Republicans still dominate a majority of the states’ legislatures and governor mansions across the country. For now, the Compact is an unrealistic option.

Another solution could be for states to adopt what I call the Maine-Nebraska Model or what is also known as the “Congressional District Method“. These two states distribute their electoral votes using the following formula:

  1. The winner of the popular vote receives the two electoral votes associated with their states’ representation in the U.S. Senate.
  2. A presidential candidate can lose the state’s popular vote, but still earn electoral votes if he or she can win a majority of the vote in each of the state’s Congressional Districts.

In 2008, Nebraska split its electoral votes, awarding John McCain four of its five electoral votes and one to Barack Obama. Obama won the majority of the vote in the state’s 2nd Congressional District, which includes Omaha and its suburbs. In 2016, Hillary Clinton won three of Maine’s four electoral vote, while Donald Trump won the electoral vote of the mostly-rural 2nd Congressional District.

What if all the states adopted this Maine-Nebraska Model? Would it have affected the results of the 2016 presidential race?

The table below includes numbers for the actual outcome of the elections, while the next column looks at the results if we apply the Maine-Nebraska Model. The table also includes the number of Congressional Districts and states each candidate won. The last column shows who won the District of Columbia’s three electoral votes.

Actual OutcomeMN_ModelStatesCongressional DistrictsDC

If all states had apportioned their electoral votes using Maine’s and Nebraska’s approach, President Trump would have still comfortably won the presidency. Should this disqualify this alternative to the EC?

Let us apply the Maine-Nebraska Model to the 2012 and the 2008 presidential elections.

Actual OutcomeMN_ModelStatesCongressional DistrictsDC

Under the Maine-Nebraska Model, President Obama would have been a one-term president. And Obama’s victory in 2008 would have been narrower.

Actual OutcomeMN_ModelStatesCongressional DistrictsDC

What if we applied the model to the 2004 and 2000 elections?

Actual OutcomeMN_ModelStatesCongressional DistrictsDC
Actual OutcomeMN_ModelStatesCongressional DistrictsDC

Under the Maine-Nebraska Model, President Bush would have earned more electoral votes in both elections.

What can we learn from this analysis? If we agree we Senator Warren’s belief that we need to change the EC in order to force presidential candidates to campaign outside “battleground states”, the Maine-Nebraska Model accomplishes this goal. After all, would Maine or Nebraska have received any attention during the last presidential elections, if they distributed their electoral votes as the other 48 states?

It is important to note that Senator Warren would probably not support the Maine-Nebraska Model. But, her desire to eliminate the EC is unrealistic. Amending the constitution will not take place in the next years.

Republicans’ opposition to reforming the EC is understandable. If it was not for the EC, Al Gore and Hillary Clinton would have won their respective races. But if public opinion starts to turn against the EC, Republicans could propose that states adopt Maine’s and Nebraska’s approach as a measure that seems more democratic than the current electoral system.

Would this be good for democracy? It would be a step in the right direction. But the chances that states will adopt Maine’s or Nebraska’s approach is also a long shot. It is not in either party’s interest. By turning battleground states into battleground districts, the parties would have to adapt their campaigning structures and raise more money to target even more districts.

For now, it seems we will just have to accept the EC’s limitations as it is unlikely that Americans will be able to reform or replace the current electoral system.

Americans’ Opinions of FEMA’s Responses to Alabama and Puerto Rico

On March 3, 2019, a series of tornadoes hit towns across Alabama, Georgia, Florida and South Carolina. The worst one decimated a number of communities around Beauregard, Alabama, claiming 23 lives and injuring close to 100.  The next morning President Trump tweeted:

FEMA has been told directly by me to give the A Plus treatment to the Great State of Alabama and the wonderful people who have been so devastated by the Tornadoes. @GovernorKayIvey, one of the best in our Country, has been so informed. She is working closely with FEMA (and me!).

Needless to say, President Trump’s words angered many. Although the Federal Emergency Management Agency (FEMA) should be impartial when it reacts to a natural disaster, the tweet suggests that the President affects the amount of assistance FEMA provides. Has Trump be playing favorites, favoring red states over blue states? Have his negative views of Puerto Rico explain why the federal government has been slow at disbursing the funds appropriated to finance the island’s recovery?

This is likely to become an issue in next year’s presidential elections. For instance, both Elizabeth Warren and Julian Castro have recently visited Puerto Rico to highlight the Trump administration’s “‘disrespectful’ treatment”  of the island’s 3.3 million U.S. citizens. Do Americans share Warren’s or Castro’s views?

The most recent Economist/YouGov Poll (March 10-12, 2019) asked its panel the following question: “Do you think the federal government was more responsive to the tornadoes in Alabama or the hurricane that struck Puerto Rico and the U.S. Virgin Islands?” 

Do you think the federal government was more responsive to the tornadoes in Alabama or the hurricane that struck Puerto Rico and the U.S. Virgin Islands_

Sadly, but not surprisingly, many Americans think that the federal government has favored Alabama over Puerto Rico and the U.S. Virgin Islands. Could this be problematic for the White House? A close look at the survey’s crosstabs show that 66% of Democrats believe the federal government favored Alabama over the other U.S. territories, while only 32% of independent and 28% of Republicans share this opinion. 

The survey also asked respondents to evaluate the President Trump’s and the federal government’s response to Puerto Rico after Hurricane Maria. This is the fifth time this survey posed this question, allowing us to track Americans’ attitudes for the past 18 months. 

Do you approve or disapprove of the way Donald Trump handled the response to Hurricane Maria_

The numbers have not fluctuated too much since YouGov asked this question in last year’s surveys. As noted above, partisanship determines how negatively respondents feel about Trump’s handling of the response. Thus, 80% of Democrats disapprove of his response, while 77% of Republicans approve of his performance. Independents, an important voting group, are more divided with 31% approving and 39% disapproving of President Trump’s actions.

Respondents were also asked to rate the federal government’s response to Hurricane Maria. 

Do you think the federal government has responded adequately to Hurricane Maria or could it be done much better_

As noted above, there is not too much change in the last few months. Republicans are more likely to have positive views of the government’s response, while Democrats are more critical. Today, 43% of independents tend to be critical, though this represents a 4% drop from September 2018.

How salient is this issue? Will it affect the 2020 presidential elections? Right now, other issues will likely overshadow the Trump administration’s treatment of Puerto Rico after Hurricane Maria. But this issue will prove be problematic with President Trump in Florida, a state he needs to win if he hopes to win reelection.

Author’s Note: A version of this post is also found in the The Puerto Rico Data Lab, a blog I created to reflect on US-Puerto Rico relations in the aftermath of Hurricane Maria. You can follow me in Twitter at the following accounts: @cyordan or @pr_datalab.

Wikipedia ‘Pageviews’ and the Early Days of the 2020 Democratic Presidential Primaries

We are just 334 days from the Iowa presidential caucuses!

So far, twelve Democrats have announced their intention to run for the Democratic Party’s presidential nomination and we are still waiting for a few more candidates, including Vice President Joe Biden and former Texas congressman Beto O’Rourke, to join the race.

Which of these candidates has generated the most buzz?

For today’s post, let’s look at Wikipedia “pageviews” data for the 10 candidates who declared their candidacy before February 28, 2019. In the past, political scientists (i.e. Smith and Gustafson 2017; and Yasseri and Bright 2016) have looked at ways these “pageviews” can help investigators predict the outcome of elections. While it is too early to predict who will win the Democratic presidential nomination, it is never to early to see how different indicators can help us think about different political events.

Wikipedia data suggests that these 10 candidates’ pages earned 4,633,621 “pageviews”. Among these candidates, Harris leads the field in terms of average daily “pageviews”.

I am surprised that Klobuchar is in second place, edging both Booker and Sanders. I am even more shocked that Buttigieg’s presidential aspirations have generated more buzz than Gillibrand’s or Castro’s candidacies.

This graph fails to take into consideration each candidate’s “pageviews” per day. For example, while Sanders may have the fourth highest average daily “pageviews”, he did not announce his second presidential run until February 19,2019 and since then many people have visited his Wikipedia page.

The following stacked barplot captures each candidates share of daily “pageviews” in relation to all candidates’ “pageviews” for February 2019. Given that Delaney’s daily “pageview” average is 28, I did it not include his data in this graph.

This plot helps us visualize Wikipedia readers’ interests on each of the nine candidacies over time. It important to note that these “pageviews” are not indication of readers’ approval for a candidates’ campaign. After all, their Wikipedia pages include information that could turn-off potential supporters. But the plot demonstrates that many people are paying attention to the ever expanding group of Democratic contenders vying for their party’s presidential nomination.


B. Smith and A. Gustafson. 2017. “Using Wikipedia to Predict Election Outcomes: Online Behavior as a Predictor of Voting,” Public Opinion Quarterly, 81(3): 714-735.

T. Yasseri and J. Bright. 2016. “Wikipedia traffic data and electoral prediction: towards theoretically informed models,” EPJ Data Science, 5(22).