Who's ripping it off? Competition theory and reality

Puzzling over the rich economies' poor productivity improvement and weak wage growth (but healthy profits), American economists are pointing the finger at reduced competition between firms. But can this explain Australia's similar story?

Jim Minifie, of the Grattan Institute, set out to answer this in his report, Competition in Australia.

Economists regard strong competition between businesses as essential to ensuring market economies function well, to the benefit of consumers and workers.

Competition is what economic theory says stops us being ripped off by the capitalists. Firms that overcharge for their products lose business to firms that undercut them.

So competition pushes prices down towards costs (which economists – but not accountants – define as including the "cost of capital", or "normal profit", the minimum rate of profit needed to induce firms to stay in the market).

Competition helps ensure that economic resources - land, labour and (physical) capital – move to the uses most valued by consumers.

Competition also encourages firms to come up with new or better products – or less costly ways of producing a product – in the hope of higher profits. But those that succeed in this soon find their competitors copying their ideas, and bidding down the price to get a bigger slice of the action.

The innovations improve the economy's productivity (output per unit of input), but competition soon takes away the higher profits, delivering them into the hands of consumers, who often get better products for lower prices.

That's the theory. Question is, to what extent does it hold in practice? And does it hold less in recent years than it used to?

The simple theory assumes any market has a large number of sellers, each too small to be able to influence the market price. In practice, however, many of our markets are dominated by two, three or four big firms.

Why? Mainly because of the presence of economies of scale. It's very common that the more you produce of something – up to a point – the less each unit costs.

So, it makes great sense to have a small number of big firms doing much of the production – provided competition ensures most of the cost saving is passed on to customers in lower prices. Which, as a general rule, it has been over the decades.

Trouble is, big firms do have some degree of control over prices. And it's common for the few big firms in an industry to come to an unspoken agreement to compete using advertising or product differentiation, but not price.

Firms can increase their pricing power by taking over their competitors to get a bigger share of the market. It's the role of "competition policy" – run in our case by the Australian Competition and Consumer Commission – to prevent overt collusion between firms, and takeovers intended to increase market power. But how well is that working?

"Natural monopolies" – where it simply wouldn't make economic sense for more than one firm to serve a particular market, such as rival sets of power lines running down a street, or two service stations in a small town - are another common departure from the theoretical model.

So, what did Minifie find in his study of competition in practice? He found evidence it had lessened in the United States, but not here.

He found plenty of markets where a few firms did most of the business. But "the market shares of large firms in concentrated sectors are not much higher in Australia than in other countries [of comparable size], and they have not grown much lately," he says.

Nor have their revenues (sales) grown faster than gross domestic product. The profitability of firms – profits relative to funds invested - hasn't risen much since 2000.

Minifie identifies eight industries characterised by natural monopoly (in descending order of size): electricity transmission and distribution, wired telecom, rail freight, airports, toll roads, water transport terminals, ports and pipelines.

Then there are nine industries where large economies of scale mean they're dominated by a few firms: supermarkets, wireless telecom, domestic airlines, then (of roughly equal size) internet service providers, pathology services, newspapers, petrol retailing, liquor retailing and diagnostic imaging.

Next are eight industries subject to heavy regulation by government: banks, residential aged care, general insurance, life insurance, taxis, pharmacies, health insurance and casinos.

(Often, these industries are heavily regulated for sound public policy reasons, but the regulation often acts as a barrier to new firms entering the market, thus allowing them to be dominated by a few firms.)

But note this: by Minifie's calculations, natural monopolies account for only about 3 per cent of "gross value added" (a variant of GDP), while high scale-economies industries account for 5 per cent and heavily regulated industries for 7 per cent.

So that means the parts of the economy where "barriers to entry" limit competitive pressure make up about 15 per cent of the economy. Then there are 29 industries with low barriers to entry making up the rest of the "non-tradables" private sector, and about half the whole economy.

That leaves the tradables sector (export and import-competing industries) accounting for 14 per cent of the economy and the public sector making up the last 20 per cent.

Even so, Minifie confirms that, in industries dominated by a few firms, many firms make "super-normal" profits – those in excess of what's needed to keep them in the industry.

By his estimates, up to half the total profits in the supermarket industry are super-normal. In banking it's about 17 per cent.

Other companies and sectors with substantial super profits include Telstra, some big-city airports, liquor retailers, internet service providers, sports betting agencies and private health insurers.

Comparing this last list with the lists of natural monopolies and heavily regulated industries suggests governments could be doing a much better job of ensuring the regulators haven't been captured by the companies being regulated.

Read more >>

Robots won't reduce the amount of work we need to do

For me, one of the most significant economic developments of this year was realising how pessimistic many of our youth have become about their prospects of ever landing a decent full-time job.

To be sure, some degree of frustration on their part is understandable. Although it's true we avoided a severe recession following the global financial crisis of 2008, it's equally true that, until recently, employment growth has been weaker than usual in the years since then.

And the burden of this weaker growth has fallen disproportionately on young people leaving education to look for their first full-time job.

What's less understandable is the way older, and supposedly more knowledgeable, people have sought to demonstrate how with-it and future-focused they are by spreading wildly exaggerated predictions about how many jobs will be taken by robots, scaring the pants off our youth and convincing them they're doomed to a life of "precarious employment" in the "gig economy".

I'm sorry to say that the otherwise-worthy Committee for Economic Development of Australia was responsible for writing on many young minds the near certitude that 40 per cent of jobs in Australia are likely to be automated in the next 10 to 15 years.

The good news, however, is that, for once, economists were moved by all the amateur analysis they were hearing to join the debate about the future of work. Dr Alexandra Heath, of the Reserve Bank, dug out the hard evidence about how the nature of work is changing and Dr David Gruen, of the Department of Prime Minister and Cabinet, put worries about the shrinking number of jobs into their historical context.

But the charge has been led by Professor Jeff Borland, of the University of Melbourne, one of our top labour-market economists.

With a colleague, Dr Michael Coelli, Borland examined the papers behind the claim of 40 per cent of jobs being lost to robots, and found it built on questionable foundations. In their figuring, the 40 per cent was likely to be nearer 9 per cent.

And last week Gruen rejoined the fray, giving a big speech about it in, of all places, Jakarta.

Predictions about what will happen in the economy can be based on the belief that it will respond to new developments in much the same way it responded in the past to similar developments, or on the belief that "this time is different".

People who know little economic history are always tempted, as many people are now, to assume this time is different.

But economists have learnt the hard way that this time is rarely very different. The fact is, people have been predicting that the latest technology would reduce the number of jobs since the Luddites at the start of the Industrial Revolution.

Gruen reminds us that, in 1953, the great Russian-American economist Wassily Leontief wrote that "labour will become less and less important ... More and more workers will be replaced by machines."

Borland notes that, in the 1960s, Lyndon Johnson established a presidential commission to investigate fears that automation was permanently reducing the amount of work available.

In 1978, Monash University held a symposium on the implications of new technologies, with the convenor predicting that, by 1988, at least a quarter of the Australian workforce would be made redundant by technological change.

Then there was Labor legend Barry Jones' prediction in his best-selling Sleepers Wake! that "in the 1980s, new technologies can decimate labour force in the goods producing sectors of the economy".

Gruen admits that "there is no doubt that, over the past two centuries, waves of technological change have eliminated jobs, and rendered some occupations obsolete.

"But they have also facilitated the creation of new jobs to take their place – either directly, or indirectly as a result of rising standards of living generating new demands."

There are two processes at work, he says. One is that technology takes jobs away – this is the bit we can all see. What we can't see is the second process, the invention of new complex tasks, leading to new jobs.

The history of technological advance over the past 200 years has shown the second process has broadly kept pace with the first.

Computers have been changing the way businesses do their business – and destroying jobs – since the early 1980s. If that's all there was to it, there ought to be far fewer jobs today.

But the number of Australians with jobs has increased by a factor of 2.7 since the mid-1960s, while the average number of hours worked per person has remained broadly stable. Fact.

Like the economists, I find it hard to believe this relationship is about to break down because "this time is different".

What's true is that the nature of work has been changing – slowly – for the past 30 years or so, and this trend is likely to continue. It may accelerate, but it hasn't yet.

Using research by Heath, Gruen says routine cognitive jobs (such as office assistants, sales agents, brokers and drivers) and routine manual jobs (factory workers, construction, mechanics) are in less demand, whereas non-routine manual jobs (nurses, waiters, security staff) and non-routine cognitive jobs (engineers, management, healthcare, designers) are in increasing demand.

It's the changing nature of work, not a fall in the amount of it, we should be preparing for.

Read more >>