Factor Markets

Derived Demand

The demand for things that go into making something is called 'derived demand'. For example, a coal mine in West Virginia has 400 employees. But natural gas prices went down by 30% over a year and a half. Because of that, power companies started using gas instead of coal, and suddenly nobody needed much coal anymore. The mine then had to let 250 miners go. Those miners didn't suddenly lose their mining skills; nobody wanted the coal they were digging up!

This whole situation is what economists refer to as derived demand. The need for any 'factor of production' - that's labour, equipment, land and so on - only exists because there's demand for the thing that the factor of production helps create.

Consider this: people want iPhones, so Apple uses Foxconn, and Foxconn hires people to assemble the phones in Shenzhen. If iPhone sales fell drastically, those assembly jobs would disappear. The company wasn't really hiring the workers as such, it was about the phones they build and whether people will buy them. Therefore, on an exam question if the demand for labour changes, you should first ask yourself what has changed in the market for the product. The boom in tech didn't mean HR departments suddenly valued computer programmers more, but that people were spending billions on apps, cloud storage and streaming.

Marginal Revenue Product

Every time a company decides to hire someone, it comes down to a simple comparison: will this next employee bring in more money than they cost?

Marginal Revenue Product (MRP) is calculated as MR x MPL (marginal revenue times marginal product). For a business in a competitive market, marginal revenue is the same as the market price, meaning MRP = P x MPL. A bakery sells loaves for $4. A new baker makes 25 extra loaves a day. MRP is therefore $4 x 25 = $100. If bakers are paid $85, you'd hire the baker and the business would make $15. But at $110, you wouldn't – it's just not profitable.

The business will continue to hire people until the MRP is equal to the wage. This is the rule for maximizing profit and it appears very frequently.

The demand curve for labour goes down as a result of diminishing returns. The third baker makes 25 loaves, the fourth 18, and the fifth 11. As each extra person is hired, the MPL gets smaller and so does the MRP. If you were to plot MRP against the number of workers, the downward sloping line would be the business's demand curve for labour. On the free response section of the AP exam, you might be given a table of production figures and asked to create the demand curve. Don't complicate things: it's just the MRP at each level of employment.

Labour Supply

Would you work a shift in a warehouse for $8 an hour? Most people would not. At $35 an hour, you might change your Saturday plans. Every hour you work is an hour you aren't having free time, and this compromise is central to labour supply.

The labour supply curve for the market as a whole slopes upwards. Higher wages will encourage more people to join the workforce and existing employees to work longer. At $12 an hour, only the most motivated people will apply. At $28 an hour, many more people will apply.

What causes the entire curve to move? Immigration is significant. A large number of qualified nurses coming to the U.S. from the Philippines will move the labour supply curve for nurses to the right at all wage levels, and will push down the average wage for nurses. Changes in what career people want to do are also important.

From around 2012 to 2022, lots more people wanted to work in tech, specifically in software, meaning there were more applicants for those positions. This is one reason why starting salaries for software developers didn't increase as quickly as those for senior engineers. Competition for jobs in general also causes changes: in 2021 JPMorgan increased the pay for starting analysts to $110,000 and as a result, fewer people applied to consulting companies like McKinsey and BCG, reducing their number of applicants.

Wage Determination

A software engineer's $155,000 salary, like the price of anything, is determined by how much of something is available and how much people want it. The wage at which supply and demand are equal is where everyone who wants to hire at that pay rate can find people, and everyone who wants a job at that pay rate can find work.

When investment in AI grew, companies rushed to find Machine Learning engineers, the value of their work (MRP) increased, and the demand for their labour increased. This pushed demand to the right, meaning wages and the number of people employed both went up. This is, in a nutshell, what happened in Silicon Valley from 2015 to 2024, and it can all be shown on a simple supply and demand graph.

The opposite is also true. If a coding bootcamp releases 10,000 new junior web developers onto the market, the supply of labour goes up. The pay for those entry-level developers goes down, but more of them find employment as companies take advantage of the cheaper workers for projects they wouldn't have staffed at the previous rate.

This is exactly the same as how things work in a typical market for goods, except the thing being bought and sold is labour, and the 'price' is the wage.

Monopsony

Now imagine a single employer in a town - a country hospital being the only place to hire nurses within fifty miles. Those nurses can't easily go and work somewhere else without moving their families, selling their homes, and changing their children's schools. This employer is a monopsony (the only buyer of labour) and has a lot more power than an employer in a competitive situation.

A normal company accepts the standard wage and hires as many people as they need at that rate. But the monopsony is the only one looking at the entire supply curve, which slopes upwards. To get one more nurse, it has to increase the wage... and not just for the new nurse, but for all the nurses it already employs! Because of this, the extra cost of hiring one more person (the marginal factor cost or MFC) is higher than the wage on the supply curve. The MFC curve is above the supply curve, and the space between them shows those additional costs of increasing everyone's pay.

The monopsony hires where the value of the extra work (MRP) equals the extra cost of hiring (MFC), then pays the wage indicated on the supply curve at that number of people. This results in fewer people being hired and lower wages, when compared to a competitive market.

And this is the part that often catches students out on the AP exam: a minimum wage set at the right level (above the monopsony wage, but below the level of a competitive market) can actually solve this problem. It makes the supply curve flat up to the minimum wage, which makes MFC fall to the same level as the wage, and the gap disappears. The company ends up employing more people at a higher wage. This is one of the few times a minimum price (a 'price floor') can raise both employment and wages, rather than creating a surplus. The exam tests this specifically because students generally think minimum wages always lead to unemployment.

Worked Example

Let's look at an example. A small furniture shop sells chairs in a market where they are $12 each, and the daily wage is $72 per worker. Should the shop hire a fifth worker?

The fifth worker can make 8 chairs per day.
The value of the chairs they make (MRP) is $12 x 8 = $96.
That's $96 in income, versus a wage cost of $72. Yes, hire. The company will gain $24 from that worker.

How about the sixth worker? At 5 chairs, Marginal Product of Labour (MPL) falls to 5, meaning $12 each for 5 of them, so $60 in total. However, the worker costs $72 to employ, so you lose $12. Therefore, you shouldn't hire them.

The best number of workers is 5. The principle is simple: continue employing people while their Marginal Revenue Product is equal to or more than their wage, and stop as soon as the next person's MRP is less than the wage. Just two figures, and one comparison.