Did you know if you add up all the imports and exports of every country on Earth, you’ll find that our planet exports more than it imports? Did you know that’s also impossible?
One nation’s exports is another nation’s imports. So for the same reason 1 - 1 = 0, there’s no way the total exports of all nations can be higher than the total imports of all nations.
However, data provided by the International Monetary Fund (IMF) shows that Earth's total number of exports is greater than the total number of imports.
So what’s the deal? How can the data defy the basic laws of economics? The simple answer is the data is just wrong.
A lot of people from countless nations and businesses contribute to the data that eventually makes its way to the IMF. Whenever someone in Britain makes a sale to someone in France, they record a value, but the person in France may not record the same value. Exchange rates, transportation costs, or poor record keeping can give multiple values for the sale.
Criminal intent leads to even more discrepancies. Someone who bought guns legally in the United States may not record their purchase when they ship it to Mexico while the arms seller would. Someone who is avoiding taxes may not record income when they move it to the Bahamas or Switzerland.
This data quality issue is not just limited to the IMF. It’s true of all data and records. From the cash registers and inventories in your local grocery store to the annual releases provided by publicly traded companies to the databases I use at work.
None of it is 100% accurate, but does that matter?
I was reminded of this question a few weeks ago when I went to a luncheon thrown by business executives at my company. They threw it to show appreciation towards myself and other business analysts who worked under them. The executive I work under was talking about his background and how he used data to guide his decisions and he said something very insightful - the data is always wrong.
He wasn’t saying it was useless. He was saying it’s never 100% accurate and it’s important to understand what makes it inaccurate so you could:
Make it more accurate, or
Understand its limitations when making decisions
No where is this more true with economic data.
Economic data tells useful trends with our economy and this data affects all of us. Politicians use this data to take credit or place blame for things they have little power over. Voters who believe these politicians will use this data to justify their vote. Business leaders and investors both here and abroad use it to assess the risk when making decisions on how to spend money here. Middle-class citizens use it to plan for retirement, buying a home, or moving.
So why would these people use data that’s inaccurate to guide their decisions? Well, what other choice do they have? Just because it’s wrong doesn’t mean it isn’t a fair representation.
What is the ultimate goal of data collection and reporting? In the end, it is an attempt to illustrate reality.
Economists to statisticians to humble business analysts like myself try to illustrate reality as best we can using data. The problem is that we can’t illustrate reality in its entirety. So what do we do? We develop tools, logic, and record keeping techniques to capture and measure as much of reality as possible.
Let’s visualize reality as this red circle:
Now imagine this blue circle as data:
Depending on how well thought out and executed the method is, we can use data to more accurately reflect this reality.
We will never be able to use data to accurately illustrate reality in its entirety, but we can get closer and closer by constantly evaluating and refining the method.
A common metric used by a wide variety of people for a wide variety of reasons is Gross Domestic Product (GDP). This metric attempts to measure how productive a country is. It attempts to add up all goods and services sold in a country’s borders.
However, this figure will never be completely accurate.
The Bureau of Economic Analysis is responsible for calculating this and they compile a wide variety of data from various sources to calculate this figure. This includes the Census Bureau, Bureau of Labor and Statistics, IRS, Treasury Department and several other smaller agencies.
This metric relies on people accurately counting and reporting sales to these departments and those departments correctly transferring the data to the BEA.
However, a lot of things are never recorded at all. Illegal drug and prostitution income won’t be recorded. Garage sales and craigslists items are likely not recorded. Productivity produced by stay-at-home spouses and family members who care for the elderly won’t be recorded either.
So GDP will never be a blue circle that reveals the entire red circle. Reality will never be fully revealed.
However, that doesn’t mean GDP is useless. If a company launches a new product and wants to market and sell it to people who can afford it, GDP is a good thing to check and verify that the people who live their can afford it.
If you were a ferrari dealer in the Middle East and you were well aware of the issues with GDP calculation, which country would you rather sell to:
Qatar: GDP per capita = $132,100
Yemem: GDP per capita = $2700
It is quite possible the true GDP per capita of Qatar is $125,000 a year and Yemen’s is $3500, but is very unlikely that Qatar’s is actually $4000 and Yemen’s is $100,000.
The same reason why investors and business leaders care about GDP is the same reason you should care about inflation data.
The Consumer Price Index (CPI) measures the prices of a basket of goods. When you read that inflation is up 2% in a newspaper, they’re usually referring to the CPI.
The Bureau of Labor and Statistics (BLS) calculates CPI. They’ll look at 175 categories they feel represents prices in America and record their average value over a year’s time.
The problems with the data usually come from their selection. For example, cable and satellite television and radio services contributed .0423% to inflation in 2014. However, that doesn’t include the prices of streaming services like Netflix and Amazon Prime. People who switch and save money don’t affect this figure.
So is it an accurate representation of prices in America? It’s not totally accurate but it’s still a fair representation.
There’s a concept in statistics called the Law of Large Numbers. Anyone who studies econometrics or statistics will have this concept drilled into their head. It states that the larger the sample gets, the closer its average will reflect the actual population.
So with 175 components and the thousands of items that are included in those components, the CPI actually has a fairly large sample. The usual recommended minimum for a sample size is 30. The CPI component for used cars and trucks alone contains 480 vehicles. So it’s a blue circle that reflects much of the red circle. It’s not entirely right, but it’s just not that wrong.
That's the thing about data. All of it is a reflection of reality, but it will never fully reveal reality. That is true with economic data and it is certainly true of your accounting and business records. However, that doesn’t mean it is useless.
Understanding and accepting its limitations will make you a better decision maker, voter, and even a retirement planner. After all, it is the same as what a famous statistician once said: “Without data, you’re just another person with an opinion.”