Why higher haircut prices might point to a strong economy (transcript and video)

Seeing the price of haircuts rise, even relative to other things I might spend my money on, is the sort of thing to make an economist rail about anti-competitive behaviour. But is that really the case, or are higher haircut prices just a sign of a strengthening New Zealand economy? Gulnara and I have a think about this in a recent video.

For those who don’t like videos, there is a transcript below.


Gulnara has told me I need to get a haircut. 

I’ve selected the place I’m going to go due to their good and reliable service.  But something I did notice is that, everywhere I looked, the cost of a haircut is a lot higher than when I was young.

Now there has been inflation – or a general increase in prices – during that time.  But this doesn’t quite cover it.  Looking at the CPI category for personal care – which includes haircuts – the prices have gone up 52% between 2006 and 2021 while the average price of consumer goods as a whole has risen 37%.


So to solve this puzzle I thought I’d call out to my favourite economist to see what she has to say.


We have all seen this graph from the United States [Vox graph].  This shows that the issue Matt is chatting about isn’t just about his haircuts – which are necessary by the way – but also refers to a wide range of different products.  

Payments for services, such as college fees and medical care stand out as items that have become relatively more expensive. 

Meanwhile durable goods, such as cars, furniture, and toys have become relatively cheaper.

As the economy has matured and grown, different needs and wants have been met, making room for different avenues for scarcity.  

Growing global trade, along with technological progress, has helped to increase productivity across a range of sectors.  However, even those sectors that have not seen productivity rise – and who have not seen relative demand change – have been influenced by this, due to the interrelationship between industries.

No matter how much we look at different sectors and industries separately, they are all bound together by a complex web of related demand for inputs – especially through shared labour and capital markets.  And it is only by considering those that we can think about Matt’s haircut.

However, answering Matt’s question doesn’t just tell us about haircuts – and it doesn’t just tell us about long-run economic history and economic development – it also helps us to understand some of the things going on, right now, in the COVID afflicted global environment.

As a result, understanding Matt’s haircut might help us to think about the risks and opportunities of the post-pandemic economy.

So let’s get this started.

Baumol effect

In the 1960s economists noticed something funny going on with the price of different things.  The 1966 paper “Performing Arts, The Economic Dilemma: a study of problems common to theater, opera, music, and dance” by Baumol and Bowen looked at this issue specifically with a Beethoven String Quartet.

Now we can all understand the beauty and value of listening to a rendition of Moonlight Sonata. And seeing such a show live is a beautiful thing.

In fact, let’s play a bit of Fur Elise.

Enough of that, back to economics. 

The quality and quantity of such performances has not really changed through time – there has been no productivity improvements in the provision of Moonlight Sonata performances.

And yet, the salary of a musical performer is now higher than it was in the time of Beethoven – in fact salaries keep rising each year (An Updated Look At Top-Tier Musician Compensation 2016 – Adaptistration).

If productivity is not rising in the classical music performance sector, why are wages rising? The key point here is that individuals have skills that are transferable between jobs – which means they can perform different tasks in jobs across the labour market. Although it is unlikely that a Cello player would go into a car factory to play cello, they would have skills to undertake some of the tasks within the factory – and could be trained on the job to do others.

As a result, if the Cello player only received the same wage as someone in the 19th century it is likely they would decide to work elsewhere – such as in the car factory where productivity has increased markedly.  

Here the ability to automate, innovate, or increase the amount of capital in “capital intensive” sectors such as the car factory, helps to boost labour productivity. Meanwhile the labour intensive “productivity fixed” nature of the orchestra does not see labour productivity rise for that job.

Wages will rise in sectors where labour productivity rises. This increase in wages will lead to upward pressure in other sectors, pushing up wages even though the labour productivity in that sector has not risen. This will in turn force firms to shut down or to increase prices – and if consumers armed with higher wages were willing to bear those prices, we could well end up in a situation where we are now.  A situation with musical performances continuing, but at a higher price and with higher wages paid.

Baumol cost “disease” refers to the Vox graph we noted earlier – it is the application of the Baumol effect to prices. However, the disease makes it sound like these higher prices are fundamentally a bad thing.  You will hear “we used to make these things more cheaply, why can’t we now!” or “it gets made so much more cheaply overseas”.  Well if it was just due to the Baumol effect, it is because the average wage rate in the economy is higher because average productivity is higher!

This is not to say there can’t be regulatory or competition related barriers that create issues.  But rising prices and wages in these “fixed productivity” sectors that are reliant on labour intensive work can just be a product of everyone benefiting from growing productivity – hardly a disease at all.



Thanks Gulnara.

On top of this, in the 1960s some macroeconomists you may have heard of – Bela Balassa and Paul Samuleson – were interested in why consumer prices tend to be higher in countries that have high average incomes, an empirical regularity that is termed the Penn Effect.

In this instance the puzzle exists because of the law of one price – the idea that in a world with open and free trade, consumer prices should be driven down to the same level across countries.

However, this need not occur, as it is costly to transport certain goods and services across borders.  Specifically, we may view some goods as “tradable” because the transport cost is sufficiently low while other goods and services are “non-tradable” because these transportation costs are high.  My potential haircut is an example of a non-tradable product, where getting a haircut in the United States would involve quite a significant amount of time and expense unrelated to the act of cutting the hair itself!

The law of one price would suggest that prices are only driven down in relation to the transportation cost – and so there should be more convergence in prices between tradable products than there is for non-tradable products.

Rising trade and increasing productivity in tradable sectors have helped to push a convergence in global prices for tradable products – and have also generated productivity increases, and in some circumstances terms of trade increases, for a variety of trading countries.  Increasing tradable sector productivity has then resulted in rising output and incomes across the world. 

But how are non-tradable prices determined? At face value non-tradable prices are determined by local demand for the product.  So if average incomes rise, due to an increase in productivity of the tradable sector, there may be increased demand for these local products which will bid up their prices.

However, as was the case with my haircut or the music discussed above, demand can only take us so far. If there is domestic competition then, even though people’s willingness to pay in the tradable sector is higher, entry to the market will see prices decline to their prior level.

Instead, the key driver of the Balassa-Samuelson effect is the Baumol effect and wages. An increase in productivity in the tradable sector translates into higher wages. Higher wages in that sector give individuals who are working in the non-tradable sector an incentive to change their jobs – driving up the wage rate in the non-tradable sector.  The higher costs in the non-tradable sector are then passed on in higher prices.

In 1989 Rudiger Dornbusch took this a step further and noted that the nature of demand changed as countries become wealthier, with demand for services increasing disproportionately EconPapers: Purchasing Power Parity (repec.org).  Such an income effect can further reinforce these changes, and reinforces the way tradable sector productivity shocks are distributed across an economy.

Combined with the Balassa-Samuelson effect, this Dornbusch result tells us that income gains associated with increases in tradable sector productivity may be weaker than suggested by GDP comparisons at market exchange rates. However, it also indicates that such gains are distributed much more widely across the economy, rather than just where the productivity increase occurs.

This can be seen in prices – as shown on this tweet.  Durable goods – which by definition are things that can be stored and so are usually tradable – have not grown in price as quickly as services for a long period of time. That is until the pandemic hit, pushing up effective transportation costs!

Timothy B. Lee on Twitter: “This might be the single most important chart for understanding the current inflation situation. For 25 years prior to 2020, the prices of durable goods like cars, washing machines, and couches fell every single year. Then in 2021 that suddenly and dramatically reversed. https://t.co/BTtqgHOKxB” / Twitter

With parts of the service sector becoming increasingly “tradable”, the Dornbusch result of rising demand for services may also not remain a driver of the Penn effect going forward – if my income goes up I do not buy more haircuts, but I might buy more subscriptions to news services. In that way, increasing divergence between tradable and non-tradable prices and their relative productivities will need to be due to the Balassa-Samuelson effect itself.

So how does this relate to Matt’s haircut? [Conclusion]

Pulling these ideas together, we have not seen great technological innovation in the type of haircut, Matt gets over his long life.  But we have seen innovations and productivity increases in tradable products, with greater productivity in New Zealand exports as well as higher relative prices received for those exports.

The productivity lift implies that there is more income overall, and part of this income is then allocated to those providing haircuts!  As a result, the lift in haircut prices is a sign of growing productivity in other parts of the economy.

It could be that there are other drivers – namely regulation or competition – but such a change could also be a symptom of a healthy economy.  Rather than a disease that we need to do something about, it is just a natural sharing of the benefits of growth.

So let’s go get Matt this haircut, and we’ll see you again next time. Byeee!