AusPol / Economics / Policy / Technology

Regulatory capture Uber alles

So the news this week is that Mike Baird’s NSW government has legalised ride sharing companies, like Uber. That’s good.


Existing taxi plate holders will be compensated for the policy change (which effectively undermines their government-granted market power) with one-off payments of $20,000 for single plates and $40,000 for multiple-plate holders. Those who bought a taxi plate in 2015 will receive the maximum compensation of $175,000 (equivalent to the current price).

But this amounts to a $250 million ‘transition fund’, ($50 million per year over five years) which will be funded by a new ‘ride sharing tax’ of $1 per fareThat’s bad.

The taxi licensing regime will continue (bad) but the price has already dropped some $50,000 and is likely to fall further. And more than 50 taxi regulations will be repealed. That’s good.

But the ride-sharing service will now be subject to a new regulator and commissioner to oversee the industry.

… That’s bad.

About a month ago Darcy called it:

New South Wales is about to set a terrible precedent if it compensates taxi license holders for their impending demise. Put simply, this is compensating failing industries for economic growth. And it is a terrible policy idea.

And then they went and did it. Clearly Mike Baird does not read Meso Soup.

In Darcy’s view, compensating losers from efficiency-enhancing policy reform is itself bad policy. I agree to a certain extent, but as an economist I’m all for exchange. While this probably sets a poor legal-political precedent, in economic terms it is unsurprising that compensation would be sought and paid.

My official position is that while (in general) compensating efficient policy reform is possible, rent seekers need to learn: don’t deal with the devil. A much better precedent would have been for the NSW government to simply legalise ride sharing with no (or much reduced, clearly inadequate) compensation. This would have sent a clear message to future players of the rent seeking game. I mean, it is ridiculous to imagine the same scenario playing out for the obsolete industries of the past. What’s changed?

But alas, the NSW government have taken the time inconsistent view. Notwithstanding the template for future bouts of rent seeking and compensation (rinse and repeat), this might still be a mutually beneficial, statically efficient outcome. Here we have competing interests bumping up against each other (i.e., the taxi cartel vs. the Uber death star platform). If an efficient compromise can be reached this means there are gains-from-policy-change, which can then be shared among parties. Everyone’s better off: the Ubers get to do their business and the taxis get compensated.

Even NSW citizens could be better off because now they have greater consumer choice at presumably lower price. That is if the NSW follows through on it’s promise to eventually drop the ride sharing tax… and if Uberisation exerts competitive pressure on the industry-regulatory complex to eventually drop the taxi licensing regime… and if the commissioner-regulator uses a light touch… and if, well, you get the point. If all this comes to pass then the approximately $250 million cost could be justified. $250 million cost and more than $250 million benefit, right?

What’s more, the $1 levy on each point-to-point trip should cover the whole $250 million, so ride sharing users should end up funding the compensation package. In the ideal world, where the total sum of compensation equals $250m and the total sum of the ride share tax take is $250m, we have (more or less) fiscal equivalence: those benefiting from the policy change are paying for it. And those who do not use ride sharing are not on the hook for any of the expense.

That’s the generous reading, it remains to be seen which way this plays out. If the actual compensation cost doesn’t equal the tax take then the difference will have to come out of (or go into) the fiscal commons, and the policy change will be redistributionary. That is, if compensation exceeds the tax take then NSW taxpayers will have to fund the shortfall (some of which do not ride share or use taxis), and ride sharers (and taxi plate holders) will benefit at their expense. And vise versa, if the tax take is too high then the excess will go into the fiscal commons: effectively a subsidy/redistribution from ride sharers to other tax payers/recipients. Let’s hope the NSW government’s forecasting is accurate.

Once more a collective action problem threatens to rear its ugly head. Those who furnish the fiscal commons are at an organisational disadvantage compared to those drawing from it. Public choice theory teaches that legislation benefiting many diffuse constituents is hard to pass because of transaction costs. Collective action is easier for the taxi industry because it is an institutionalised cartel, and it’s not-too-difficult for Uber either because it’s a solidary organisation. If the ride sharing tax take does not end up equal to the cost of the compensation package then the taxi cartel and ride sharing companies will benefit at the expense of NSW citizens: either the group of (non-ride-sharing, non-taxi-using) taxpayers or the group of ride-sharers. In either case, for members of either group their share of the difference (spread over a long time horizon) is going to be negligible, probably in the order of a couple extra tax dollars per year. This is a classic formulation of diffuse costs and concentrated benefits that public choice theory warns of. They never stood a chance.

Moving on. My biggest issue with the NSW ride sharing decision is not the compensation, nor even the “short term levy”, but the spectre of regulatory capture. The key part of the decision is the new regulator and commissioner to oversee the ride sharing industry. The sharing economy is already threatened by over-regulationYes there is a precedent and template for future bouts of compensation (when ride sharing is one day surpassed) but the taxi cartel also provides a very successful template for regulatory rent seeking. The threat is that Uber simply comes to replace the taxi cartel: it won’t be long before they’re writing their own laws, locking out current competitors or future personal transportation business models (perhaps around the technology of driverless vehicles). As Stephen Livera put it: “either you die a hero or you live long enough to become a villain” (you should see his post on the decision too).

As Darcy wrote in the latest IPA Review, sharing economy companies like Uber and airbnb pressure governments to permit new, more efficient markets to emerge and supercede inefficient, protected old market models:

This locking of horns between the incumbents and the innovators is a good thing. Each new user of the sharing economy is a vote for the free market, and is additional power on the side of the free market political entrepreneurs.

But as I noted here on Meso Soup:

Lest we forget that today’s free market political entrepreneurs too often become tomorrow’s special interest political pleaders. Such is politics.

While it’s not unreasonable to suggest some small, temporary market power (inefficiency) could be exchanged for the long-lasting, more efficient institutional innovation, we must be ever vigilant that the Ubers and the airbnbs don’t become the entrenched special interests they replaced. One way to do this is to make regulation technology-neutral to avoid entrenching industry structure:

Technology-specific regulation only survives the test of time when there is little innovation. Yet traditional industry structures are continually being displaced. Creative destruction is a good thing.

However, when governments regulate an industry, these regulations by their nature define and determine the structure of the industry.

Many sharing economy regulatory contests come down to questions such as ‘what is a taxi?’ or ‘what is a bank?’ As industries shift and innovate, these definitions blur. But regulatory frameworks tend to be fixed, based on the assumptions built into the industry structure that they were original designed to govern.

If governments want to encourage the sharing economy, they need provide a reliable, predictable, technologically-neutral legal system that both keeps industry-specific regulation to a minimum and favours private solutions to regulatory problems over public ones.

Such an approach will forestall any attempts by today’s disrupters to become tomorrow’s special interest political pleaders via regulatory capture. Then the transition from ride sharing to driverless taxis might be less fraught than the current taxi-to-ride-sharing transition.

The other great hope is that decentralised, crypto-alternatives to Uber (such as La’Zooz) begin to emerge:

Using the same technology underlying the virtual currency Bitcoin—a distributed online ledger, or “blockchain”—the La’Zooz network would exist on the phones and computers of its community of users, rather than any central server.

As with other crypto-technologies, rent extraction via regulation and taxation would become much more difficult (if not impossible). I’ll end with the prospect of pairing much lauded Ethereum decentralised autonomous organisations with equally lauded Google autonomous vehicles. Regulate that.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s