Tuesday, March 20, 2018

Unit Labor Costs

One of the creepy things about learning economics is all the technical jargon. Diminishing marginal utility, gross private domestic investment, and the production possibilities frontier are good examples. What the hell are these things? That can be the subject of another post because today I want to focus on another term economists throw around like salmon at the Seattle Fish Market.

Today I want to discuss unit labor costs. Say that 30 times and I promise you and all those within 50 feet of you the best sleep you have had in months. It's better than a My Pillow. To make sure I don’t lose you, let’s rename it ULC. ULC rhymes with sulk but that has no consequence here. 

ULC could be the most important macroeconomic variable in town this year. So you should know her. If you saw ULC sitting alone at the end of a bar, you would ordinarily tip toe quietly in the other direction. But this year, ULC is the Queen of the Ball.

That’s quite a claim. Yet I don’t hear anyone talking about her. Before I am done with you today, I want to convince you that ULC is at the heart of many issues we discuss today – rising wages, rising prices, the next recession, and of course, the size of our new tariffs on Armagnac.

Imagine any product – let’s think about a bottle of JD. Most of the cost of producing one more bottle of JD is what you pay the employees to produce it. This includes their wages and any other earnings they might receive in the way of benefits. Logic suggests that, if everything else is the same, a rise in labor costs means the company will have to charge more for a bottle of JD. If it used to cost $10 to produce another bottle of JD and now it costs $12, then one would expect the price to reflect that cost increase. 

The labor representative will interrupt us now and point to the fact that the extra wage won't lead to a price increase because the JD workers were more productive this year. The price of a bottle of JD depends on the labor cost and the labor productivity.

For example, suppose wages go up 5% this year. Don’t we have to charge more for a bottle of Jack? Nope – it depends on how productive labor was. Suppose workers got really jacked up this year and produced 10% more bottles of Jack. That is, you got 10% more Jack for 5% more money. In that example the cost of labor in each bottle of Jack was lower! Thus they can sell the Jack at a lower price.

ULC is, therefore, a delicious macro concept that summarizes the key factors impacting the cost of goods and service:
  • ULC went up – cost per unit is higher and therefore we ought to raise price (or take lower profits)  
  • ULC went down – cost per unit is lower and therefore we can lower price and be more competitive (or keep prices the same and take higher profits)
  • ULC did not change – cost per unit is the same. Go fishing.
So you can see that ULC is a vital part of the economy and yet most of you thought it meant Underware Latex Creep.  

Here is the most fun part. The graph at the bottom of this life-saving exercise shows you the history of changes in ULC since before Joe Biden was born. I won’t bore you millennials with all that history stuff but you can see that before 1980, ULC was quite the party animal. It was all over the place – rising by almost 13% in 1974 not long after a mere blip of 1% a few years before. Imagine if you were selling Kool-Aid in the front yard and the cost of getting your mom to make the Kool-Aid for you rose by 13% one year. If you passed that cost increase on to your customers, they might decide to go down the road to Peter's house.

Notice that since 1980, ULC became more well-behaved. It has its cycles but they are much less pronounced. The mean change of costs per unit is about 2.5%. Notice also that just about every recession —the grey bars – was preceded by a rising trend in ULC. Clearly, when costs per unit are on the rise the resulting higher prices of goods and services seems juicy – but if this keeps up the economy can no longer handle it.

This brings us to the present. The average change in ULC is remarkably less than 2.5%, and there is no discernible upward trend. After it dropped into negative territory recently, it jumped back to positive territory but there is no clear upward trend. In fact, if you look at behavior since around 2011, you might see a downward trend.

If it ordinarily takes a few years of rising ULC to cause a recession, there is little in this graph to suggest an imminent recession or slowdown in the economy. But aren’t wages beginning to rise faster? Won’t that make ULC jump and signal bad times ahead? Yes, wages might rise but remember ULC is impacted by productivity changes as well. If productivity changes as much or more than ULC – then ULC won’t change at all.

So fasten your seatbelts, kids. This is a race between wage growth and productivity. Which one are you betting on for the next few years?


  1. Dear LSD. Talking heads chatter about technology improving productivity and its impact on keeping ULC down, thus off-setting the inflationary effect of increases in dollar wages. Also, the increase in labor-saving-designed capital investment—e.g. plant and equipment aside from computers, AI, internet, etc.—will further reduce ULC. As the economy picks up steam job creation will demand more bodies, but that inflationary effect will be off-set to some degree by more bodies re-entering the work force. Further, immigration (legal and illegal) will keep dollar wages down. I think wages and productivity will fight until the 15th round to a draw.

    1. Thanks Tuna. Looks like we agree in the sense that we don't see ULC as being a big threat at the moment. If rising uLC is often behind recessions then maybe this is one positive factor in the outlook.

  2. LD--
    Can you comment on what might be the long term effect on this measure as the economy moves away from the "production of a lot of things( e.g. bottles of JD) by a lot of people" to products/solutions (think: software apps) which are increasingly being created, made, and distributed by one, or a handful, of humans? It seems to me that the "productivity" component of your equation might be significantly different.

    1. Nice point Ed. Keep in mind that we have been moving away from a classical goods economy for along time. Consumer services are about 70% of all consumer spending, for example. Yet ULC seems to still be important.

      As less and less human input goes into production, this raises productivity. Of course if labor is replaced by other variable inputs then we might want to replace ULC with the more general unit costs or UC. The same theory underlies this indicator but it will be tougher to measure than ULC -- since the variable input would have to be measured.

      If that is not challenging enough there is the issue of whether our usual notions about productivity hold in a world of non-labor variable inputs. Maybe in that world more inputs are not subject to diminishing returns? For the present I will stick with ULC as being useful -- but I recognize that as the nature of production, inputs, and productivity change, we may have to find a better indicator.