Flying blind in data collection, VI could use an institute for statistical analysis

[view original post]

Data is vital to making choices and decisions at all levels — individual, business, and government. Because we live in a world of unlimited wants and scarce means, choices and trade-offs must be made, and sound analysis of data helps make optimal decisions.

Residents, for example, may have to choose between eating at home or eating away from home in order to save money toward a future down payment on a car or house. Business owners may have to choose between leasing or purchasing a piece of equipment. Prospective external investors to the territory would need to know the cost of living, the ease of enforcing contracts, the cost of energy and water, the frequency of electrical interruptions, average commercial rent, the distribution of income to know what to pay employees, how to price goods or services, whether to invest in a generator or solar panels and still earn a return.

As government officials there would be a need to analyze how much tax incentives to give away to big investors in exchange for job creation, and the hope of higher income tax returns. In all of these cases what will be needed are accurate, reliable data on prices, interest rates, inflation, expected inflation, expected growth in sales, expected profitability, the value of foregone revenue, the expected economic impact of the investment, and the size of multipliers between different sectors of the economy to capture the ripple effects of backward and forward linkages.

Data, however, is scarce, unreliable, and incomplete in the Virgin Islands. Vital economic and social data are not systematically gathered, cleaned, made publicly available and rarely analyzed. To often, decisions are being made with outdated, incomplete and inaccurate data, leading to bad outcomes. We, too often, have to accept the data of projections of others who have vested interests. We suffer because we do not have the correct data and analytical capabilities for our proper due diligence (the VITOL propane conversion deal comes to mind). We fly blind in negotiations.

We, too often, cannot quantify the dimensions of a social or economic problem to compete for federal grants successfully. And, because we do not have a culture of monitoring and evaluation, we cannot easily measure and evaluate performance, and thus do not know if a program is effective. We fly blind in begging Uncle Sam for help, and thereby, lose some opportunities, and waste many opportunities we have landed.

Currently, the Virgin Islands does not appear in many leading databases, whether free access or paid subscriptions.

Locally, there are six principal sources of economic and social data. Still, only three have web-accessible data, and two of the three present data in downloadable form (non-pdf), so only two are helpful to a researcher.

• V.I. Labor Department

• Bureau of Economic Research

• Education Department

• Health Department

• V.I. Port Authority/WICO

• Tourism Department

Moreover, there is a glaring lack of data on banking, the environment, and businesses in the Virgin Islands. We do not know how much credit is going to different types of businesses. We do not have data on the functioning critical ecosystems, or data on which companies are growing, failing, surviving, increasing productivity and profitability, losing productivity, and profitability. Thus, we do not know leading sectors or lagging sectors.

In social data, we have to wait for the decennial censuses and the Virgin Islands community surveys to find out about poverty, the structure of families, and population demographics. The Virgin Islands has no clear idea of its income distribution. No Gini Coefficients have been calculated.

From a macroeconomic perspective, a set of critical variables need to be known and monitored all the time — real economic growth rates; projected growth rates; unemployment rates; inflation, interest rates; credit supply; labor force participation, poverty rates, income inequality, ease of doing business indices; debt- to Gross Domestic Product ratio; revenue as a share of GDP; actual vs. projected revenues; projected expenditure rates vs. actual; debt service as a share total public expenditures; and productivity. We do not have a macroeconomic model of the economy and cannot do simulations of the likely impacts of changes in taxes, for example. But yet we blithely consider sales and internet taxes because we need more revenue. Without these numbers on a dashboard and basic stimulation models, we largely fly blind in the areas of financial management and economic policy making.

In the area of educational data, there is a wealth of administrative data, but it is not thoroughly analyzed. Currently, we are amidst a COVID pandemic. There is no mention of what disparities and achievement gaps may have appeared between public school students with internet connectivity at home and those without internet connectivity and the impact of the delays in handing out laptops and hotspot connections. We run the risk of an entire cohort of students being adversely affected by two closely spaced external shocks — hurricanes Irma and Maria and COVID. We don’t know the degree of learning deficits and, subsequently, the best targeted remedial interventions. We fly blind when it comes to educating our children.

The best step to improve statistical and analytical capacity would be to create a statistical and planning institute with the following characteristics and mandates.

First, the institute should have a legal mandate to collect social and economic data periodically from public sector agencies and private businesses. The data collection process should be routinized and automatic as much as possible. Periodic surveys of households and enterprises need to be planned and adequately funded.

Second, a governance system must be designed to address confidentiality, privacy, and security issues.

Third, the institute should be well-funded and staffed with adequately trained people. The institute should attract persons trained in mathematics, statistics, computer programming, data science, population demographics, economics, psychology, and sociology.

Fourth, the institute needs to be independent and adhere to the highest standards of integrity and transparency. It should be run by technocrats committed to using the best methodologies available, publishing data periodically, and engaging in objective analysis. Its quantitative social and economic reports should be nonpartisan and objective.

Additionally, the institute should publish reports without the review or veto of the chief executive. Political appointees will be too tempted to issue only statistical reports favorable to the standing administration. The institute should consist of most community members, the business sector, and academics, not government officials. The director should be appointed to terms asynchronous with the gubernatorial election cycle. Given the long history of cronyism and favoritism in V.I. politics, independence of the institute should be a necessary and nonnegotiable condition.

Fifth, the institute should sell services to the private sector and actively pursue grants to diversify income streams and insulate itself as best as possible from local budgetary politics.

Precedent exists in other insular territories for the creation of statistical bureaus and institutes — Guam (2002) and Puerto Rico (2003).