It would be safe to say that everyone has likely heard the term 'economics' and even probable that most people have an idea of what economics is all about. Most would say it's about money and putting it to work for maximum benefit. To an extent, they wouldn't be half wrong.
The question is: which benefit? Whose benefit should be maximised? And what defines 'maximum' in the context of economic benefit, especially if it is not yet known if maximising a benefit might mean minimising a different economic advantage? Or worse: harming a section of the population or the environment.
Is all of that clear as mud on a sunny day?
Let's take a common money question as an example: should one put money into savings when s/he has debts to pay?
Saving money and planning for your future financial security is always a good idea. But if you owe money, your lenders will charge you interest, which would likely be far greater than any interest you could earn on your savings, making savings less beneficial to you. And, the longer that debt remains active, the more 'harm' you incur.
And then, consider that your bank will make more money than you do off of any savings you build, effectively reducing your benefit and increasing theirs.
Simply looking at who benefits more if you save money (lenders and banks) rather than eliminating debt (you) doesn't answer any of the bigger economic questions.
What if you suddenly need the money would have had in your savings account? What if you suddenly inherit a windfall, making investing, rather than saving, possible? What if you refinance your debt at a lower interest rate - does that decrease your 'harm'? How much would your saved money be worth in a few years if you factor in inflation?
That's where econometrics comes in. This particular brand of economic research and analysis uses statistics and analytical tools to look at measured economic phenomena and provide possible clues policymakers can follow to guide fiscal policy.
To do that well, econometrists aim to achieve three goals. Your Superprof breaks them down for you now.
Goal Number 1: Analysis
Disclaimer: the example above is not suited to econometric analysis unless the econometrist examining those economic phenomena look at savings and borrowing trends across a population. And, even then, they would not make recommendations for individuals to follow but for policymakers - perhaps to raise or lower interest rates.
It's impossible to formulate a statistic from a single data point.
Econometrics cannot apply to individual situations because the discipline draws on statistics as well as economics, and involves the use of mathematics. However, that's just the type of information that econometrists consider when tasked with proving an economic theory.
This process starts when a specific economic phenomenon comes under consideration.
Throughout this article, we'll use the wage gap between workers with and without a university degree in a particular sector - maybe manufacturing or marketing and sales to illustrate salient points.
To make the analysis more specific, the econometric theory written to address this question might specify to only consider workers of a certain age or level of experience in that field. That's because time on the job, particularly with the same company, is a variable that could influence workers' wages, too.
Side note: find out how large a role theoretical econometrics plays in shaping economic policy worldwide...
To start the analysis process, econometrists gather all of the relevant empirical evidence that tests the economic theory in question. Does the observed economic behaviour - the empirical evidence support the economic theory?
More often than not, econometrists end up with a 'sliding scale' result. To reflect such an outcome in relatable terms, now...
They will likely find that not every single worker within that that certain age group and/or level of experience will earn the same amount. For instance, workers living in big cities may earn more simply because their cost of living is higher and, thus, they get paid more.
That is a variable that was not considered in our example; another is the size of the company they work for. Multinational companies can and usually do pay their workers more than smaller, homegrown concerns.
Still, all other variables aside, after examining all of that data, a clear picture emerges of whether or not workers with a university degree earn more.
That conclusion could have a wide range of consequences. Governments may make university education more affordable or provide grants and subsidies to those who cannot afford to continue learning. They may raise minimum wage, forcing employers to increase skilled workers' pay to maintain the gap between skilled and unskilled labour.
Goal Number 2: Supplying Estimates
Some people might play around with linear regression just for the fun of it but most econometrists use the multiple linear regression model because they have been tasked to find information that addresses a specific economic concern.
Usually, the bodies assigning those tasks are the same ones that set fiscal and economic policy for their country or group of countries.
After econometrists have modelled all of the data applicable to the econometric theory they were given, they have to do something with the conclusions they've drawn. They can't simply throw a bunch of scatter plots on the table and claim their work is done; they have to explain their conclusions and provide concrete numbers that will help shape economic policy.
That's why, before presenting their analyses, econometrists winnow their findings down to the coefficients of economic relationships, a simple numerical estimate that policymakers can use to decide how they'll move forward armed with the information at hand.
Econometrists arrive at that number by using correlation coefficients to measure the strength of two variables' linear relationship.
You may be familiar with the terms 'positive correlation' and its inverse, 'negative correlation'.
If the correlation coefficient is more than zero, that number signals a positive correlation between the two variables. If the correlation coefficient is less than zero, their relationship is negative and if the correlation is zero - neither positive nor negative, there is no relationship between the two variables.
Applying that knowledge to our wage gap hypothesis, now:
- the correlation coefficient is a positive number: wages increase the higher the education level
- the correlation coefficient is a negative number: the lower the education level, the higher the wage and vice versa
- the correlation coefficient is 0: the level of education has no impact on wages and wages do not reflect workers' education levels
You might wonder how workers with a lower level of education could out-earn a university-educated worker. In fact, that happens much more often than you might think!
Picture an entry-level manager with no university degree who, by virtue of a promotion, has moved from hourly wages to a salaried position. Unlike wager earners do, s/he won't earn any overtime and, because of that entry-level status, may not yet qualify for any bonuses.
Of course, that is a vastly oversimplified example - suitable for this introduction to econometrics, but it does show how wage earners can out-earn management workers.
And, overall, it gives a fair snapshot of how econometrists turn raw data into concrete numbers through quantitative analysis.
Goal Number 3: Forecasting
After all of that modelling and analysis, you might think the econometrist's job is finished... but it isn't, yet. Now, they have to forecast what will happen if the current statistics continue in play and what policymakers should change to improve the economic outlook. How do they do that?
Let's say there's a positive correlation between university education and wages in that particular industry. The logical conclusion is that, if more workers had a degree, they would earn more - meaning they would also spend (consume) more and pay more in taxes.
Those are two main reasons why policymakers consider this scenario, by the way.
In forecasting, econometrists consider risks that might impact the prosperity of an economic phenomenon, usually up to a set date in the future - maybe one year or five years, but often at smaller intervals.
In our wage gap scenario, such risks might include a percentage of higher-paid employees finding work in other fields or retiring, less demand for employees in that field because that market is stagnant and how automation might reduce the need for workers.
They select the variables most relevant to the economic question, develop statistical models to predict how those variables might behave.
Reaching the three goals of econometrics - analysing, estimating and forecasting, is exacting and demanding.
Many entities concerned with economic policy rely on econometrists to achieve these goals so those policymakers can determine which direction to take in setting their economic goals and boosting economic performance.
Those entities include the World Bank, the International Monetary Fund, and, of course, the Organisation for Economic Cooperation and Development or OECD.
With that being said, we have to touch on the global economic downturn of 2008, an event that few forecasters saw coming. With all of that data scrutiny, how could they have missed the warning signs?
Economists and econometrists struggle to explain that still today. And not just that event but the fallout from it, too, the effects of which are still being felt around the world. Now add to those residual effects to how the pandemic upended global economics...
In light of that revelation, do you think econometrics should be accorded the importance it currently has?