Why is important data inaccurate?

Wanlikhang

20 April, 2023

#International

This is the second article in our layperson’s series How Economic Data Deceives. In our first article, we described some of the inaccuracies:

Vested Interests - It may suit the agendas of those commissioning, preparing, and revealing economic information to deceive us for their own political, commercial, or personal ends.
Delays - Slow collection and analysis often delay information until is useless. This can be months or years. Estimates or wild guesses are made. These allow further manipulation.
International institutions create different numbers - The World Bank, the OECD, the IMF, and others have the same overarching concepts. Countries supplying them with information use widely different methods. Those publishing the information make estimates, manipulate, and interpret the inputs. Their statistics for individual countries can differ to the extent that they are misleading for international comparisons.
Sloppy and unqualified data collection - There are relatively few sophisticated, developed countries. The process of collection is questionable for much of the world.
Dubious measurements - Many economists argue that important statistics, especially Gross Domestic Product, GDP, are unsound. When doubtful numbers are used with others to calculate productivity or growth the results are grossly misleading. We will expand on this below.
Humans tend to use bad numbers when no others are available – This can lead to implementing wrong policies.

But why is this important data always inaccurate?

Methods of collection

Absolute rulers simply publish the results by which they wish to be judged. In democracies, common sources of conscious or unconscious bias include:

Easy to collect information (from income, VAT or import tax submissions, banking statistics, census records for example) can be wrong. Tax evasion and fraud are almost national sports in many countries. In poor countries, the unbanked often form the majority.
Surveys are used to calculate other information. Those designing and collating survey data have many ways to distort or manipulate the data. For example:
- newspapers conduct surveys of readers’ opinions. It is hardly surprising that they reflect the politics and prejudices of readers rather than the entire population.
- The wording and order of questions can suggest a desired response.
- Varying response rates and sample changes over time make tracking changes difficult.
- Paymasters influence the results. A sophisticated European country has a government-controlled unit to conduct official surveys and present the results. It also conducts surveys and creates further analyses when paid by third parties to do so. An analyst who works there explained: “We ask what they want to prove and ensure that they get what they pay for.”

Unsound GDP concepts

Politicians, businesses, charities, the media, and others quote Gross Domestic Product figures daily to support their points of view. Common objectives may be to prove: Why capitalism is better or worse than socialism; why being in or out of the European Union makes sense; how much better country A is doing than B; or how a particular government is performing.

GDP is defined as all consumer spending added to business investment, all government spending, and net exports. (Exports minus imports). It purports to be a measure of overall economic performance. However:

Depleting natural resources is not included. Countries increase GDP by mining, oil extraction, deforestation and other measures that might harm the environment. Coal mining and fracking, for example, contribute to US GDP.
Depending on culture, degree of sophistication and more, not all consumption is recorded. Reasons for this might include bartering; avoidance of sales taxes; spending on illegal substances; or perhaps purchases in remote locations unknown to the authorities.
Businesses have incentives to misclassify investment. In many countries, business investment is deductible from profit for tax purposes. This encourages over-reporting of investment. An investment bank I was with was asked to sell an Italian corporation. In this case, the business had no apparent profit, because it falsely charged its construction of an enormous plant as the cost of goods sold. This also meant that the balance sheet showed very little in fixed assets. We were astonished by the ultra-modern and huge factory. It was difficult to explain this to prospective buyers from more conservative countries. Potential buyers were being asked to pay for a business that had no apparent profit or assets.
Government spending can be good for the welfare of the people - or not. Officially, US spending on defence in 2023 will be $1.73 trillion. This does not include ‘Black Programs’, those funded from unofficial or secret sources. Views vary as to whether this level of spending contributes to national well-being.
The effectiveness of state spending varies. French and Chinese railways are more cost-effective than British ones. Singapore’s state-operated, airline, port, power generation and social housing compare with the best-run private enterprise operations internationally. Most state-owned enterprises do not.
The nature of trade is intriguing. Mexico, Colombia, Peru, Ecuador, and most of Central America are notorious for the traffic of illegal drugs. Online publication qcostarica.com attributed much of the recent strength of Costa Rica’s currency to drug cartel dollars passing through.
Work conducted on one’s own behalf is excluded from GDP figures, as too difficult to measure. This includes household chores, child rearing and more in the UK and USA. In poorer countries, it can dominate economic activity, for example building one’s own house or growing most of the food on a smallholding.
The case of the Irish Republic deserves special mention. Over several decades, Ireland was able to attract inward investment by offering lower corporation taxes than other European Union countries. This led to major corporations investing in Ireland. Many paid their corporation taxes there by channelling royalty payments and profits through their Irish entities. In 2015 there was a single-year spike in the Irish GDP of an incredible 32%. Investigation indicated that major US technology corporations chose to realize their profits through the country in that year. There was no obvious benefit to Ireland of this huge increase in GDP,
The value of GDP does not allow for inflation without needed adjustments. By ignoring inflation, a country can claim it has grown and avoided recession when its economy has shrunk if inflation is considered. Arbitrary and complex adjustments to nominal GDP attempt to allow for such problems. Purchasing Power Parity, PPP tries to show what money can buy in a country. As an example, both China and the US spend a lot on defence. A Chinese tank or fighter aircraft costs a fraction of one in the US. The difficulties of estimating the technical and quality differences between these items are legion, with partisan arguments on both sides.
Apologists for the OECD, IMF and World Bank claim some commonality of approach and therefore some validity to the GDP data they publish. Given that all three are heavily dependent on US support and subject to political appointments, this assertion should be challenged.

GDP per capita

GDP forms the basis for another measure of economic success. Dividing the output of an economy, (GDP), by the number of people living in a country provides data on income per head.

It sounds simple enough.

However, how is income distributed across the population? Are a few families rich, as was true of President Suharto’s Indonesia, and Marcos’s Philippines, or is the wealth spread around? Finland and Slovenia are often given as examples. Ever-inventive economists have developed ways to try and measure wealth inequality. Corrado Gini invented the Gini coefficient over 50 years ago. It measures the spread of national income across the population. Unfortunately, it relies on the doubtful accuracy of GDP data.

Identifying how many people live in a country is critically important for GDP per capita and the Gini Coefficient. This sounds simple but adds further inaccuracy to economic statistics.

It is hard, but surprisingly common, to live outside the bureaucracy in sophisticated countries.

Illegal immigrants and criminals may choose to do so.
There are countries with remote populations, rebel movements and poor records of births and deaths. India is the largest one.

India has surpassed China in population. Yet the Indian population is difficult to determine. India, a country the size of Europe, with 21 recognized languages and high corruption, began its last census in 2010. The next one has been delayed until at least 2024.

With India’s fecundity and poor general education standards, the current population estimate must be wildly speculative. As a result of all we have discussed, estimates of India’s GDP per capita, growth and Gini coefficient seem meaningless.

Enter the Spooks

Those of us who needed to make international comparisons for business analysis became increasingly frustrated by the lateness and inaccuracies of published data. The CIA’s world ‘factbook’ is widely used.

This purports to compare countries in multiple dimensions. Besides data from the IMF, OECD and World Bank the CIA has unique access to satellite images. They can show harvest yields, view the details of natural resource extraction and transportation. Secret viewing of the internal documents of other countries and corporations is also available. Other nations deploy their own clandestine services for similar purposes. Unfortunately, partisan motives make all such ‘facts’ of dubious value.

Let there be Light!

The quality of international measures of performance is so bad that some have proposed radical improvements. Using light emissions from a country is interesting. Many readers will have seen satellite pictures of poor North Korea in darkness compared to prosperous South Korea’s night-time lights.

But there are many problems with this approach also:

Culturally some countries use more lights than others. Paris has lower light emissions than Shanghai, yet the wealth of Paris is much higher.
The Leninist idea of focusing on electricity production for heavy manufacturing rather than domestic consumption still influences some economies.
Though unlikely, perhaps Kim Yong-un has switched to being green.

Conclusions

The main methods used to measure economic performance are deeply flawed. They become more so when countries are compared.
GDP and its manipulations are widely used as a basis for absolute performance by a country. Overlaid with dubious population numbers to estimate average wealth in an economy creates further distortions.
Variations in GDP over time are used to measure economic growth, an obsession among many. The same objections to compounding errors apply.

Future articles

Trade between nations has been a major source of world development. In recent times, sanctions, wars, and nationalistic policies have disrupted the flow of goods and services. We will explore the problems of information on trade and currency flows.

Nobel laureates and famous economists have tried to resolve the problems of false performance measures. They have challenged the obsession with output, growth, and productivity. Various ways of measuring ‘happiness’ and ‘well-being’ are now in vogue. We will review the problems they present.

About the author:

Chris Clarke is a friend and former colleague. He has degrees in economics and management from the UK. He is a past chairman of the Strategic Planning Society. Working around the globe, he held leadership and partnership positions in strategic consulting, corporate finance, and advisory services. He was based in Europe, Asia, and the US, before moving to Central America. On a pro bono basis, Chris was a visiting professor at Henley Management College. He taught senior executives in Asia and Europe. For some years he also ran courses in M&A and corporate strategy for Management Centre Europe.