Tag Archives: Dynamic pricing

Uber surge pricing no substitute for operations planning

I got the this text from Uber yesterday and the graphic pasted below showing an hour by hour prediction of demand for New Years Eve.

UBER: 3 days until NYE – the BUSIEST day of the year is almost here! You can make great money driving between 7-11pm and 12-6am on New Years Day. See you on the road!

Your Uber Team

Unfortunately for Uber these texts promising riches are losing their effectiveness because they are no longer credible. It was the same for the Rugby World Cup – as soon as a text went out promising windfalls at Twickenham I just left the area.

Of course it makes sense for Uber to flag up periods of high demand and encourage drivers out onto the road to meet the demand. But typically this results in market over supply with meagre money for the many and much money for Uber. Last year a colleague made £90 gross working 12 hours on New Year’s Eve for Uber and that was typical. I suspect Uber knows this well – it certainly has the data to divine this – but chooses to mislead drivers to maximize its own profits.

So what?

What this proves is that surge pricing, at any time, is totally unnecessary and represents only a failure in operational planning. The experience I and other drivers have had with depressed incomes when Uber promises high earnings shows that Uber has tremendous capability in demand prediction and supply side mobilization. Some of us have also anecdotally noticed an up tick in surge pricing towards the end of  each quarter, perhaps due to internal pressure to meet revenue targets.

Uber NYE

Given Uber’s superior ability to predict both demand and supply so well then surely setting modest surge price increases in advance would likely clear the market more efficiently and at less cost than allowing the algorithm to try to solve the problem dynamically when there is simply no available additional supply in the market to instantly tap? At any rate, the market cannot operate effectively in the long run so long as suppliers do not have the information they need to make a decision to participate.  Perfect markets operate effectively when all actors have all information. When demand and supply information is hidden under the table by a market dis-intermediator such as Uber then efficient outcomes cannot be achieved. Taking Uber’s word for it when earnings fail to pan out time after time won’t wash for long.

If Uber keep crying wolf to drivers with promises of earnings that don’t materialize then it will have no choice but to either offer realistic incentives in advance to get drivers on the road or clobber riders with run away dynamic surge increases which just might tempt the patience of the regulator or the riding public.

The invisible hand is no substitute for the visible hand of effective operations management.

Did Uber price gouge during the London tube strike?

Wednesday night and all day Thursday was crazy for us London Uber drivers and even more so for commuters. Tube strikes on one line or another are fairly frequent but this is the first time I’ve seen the whole system be taken down in a strike. The price up lift of 2.9x was a welcome boost to earnings but the gridlock in London made it difficult to turn over the work at pace. One journey from Picadilly to Old Street took nearly 90 minutes and cost my poor passenger £50. Ouch!

Admittedly, most passengers took the surge pricing in their stride and mostly seemed philosophical about it. Maybe the ones who were truly indignant refused to book and those who couldn’t afford it walked home. Spare a thought for London’s key workers on modest wages forced to live far out in the suburbs due to high housing costs.

As I sat in the traffic I had time to think about the system. Was I instrumental in providing an essential service efficiently rationed to the highest bidder according to economic laws? Or was I part of an opportunistic scam to unethically price gouge Londoners in their hour of greatest need?

The Telegraph was perhaps alone in making the case in the media that Uber was right to triple its fares during London’s tube strike. The reader poll with the article shows 75% of readers agreed with the proposition. The article drew on a poll of so called influential American economists who wholeheartedly agreed with the proposition using surge pricing to allocate transportation services – such as Uber does with its cars – raised consumer welfare through various channels, such as increasing the supply of those services, allocating them to people who desire them the most, and reducing search and queuing costs.

But the economists did have some important caveats to their opinions. One pointed out that consumer welfare and market efficiency are not the same thing. Another agreed with the proviso that there are no major externalities. (Many might argue that the strike and associated grid lock was indeed a major externality.) Another agreed provided there was true competition in the market place. Uber’s UK MD, Jo Bertram, provided an interesting analysis of competitor performance during the strike relative to Uber’s.  Yet another economist ceded the point but pointed out that efficiency was just one outcome and anti gouging laws protected people from market efficiencies that have anti social effects. For me, the real insight came from one economist who agreed with the proposition provided the surge really did incentify additional supply into the market otherwise it would be a straight transfer to Uber.

In comparison, The Telegraph’s argument is facile. Peter Spence argues that 2.9x prices helps Uber to manage digital queues that would become over whelmed by rampant demand. I could see this might be the case if a local minicab office despatcher suddenly had to take 50,000 calls but Uber’s e-hailing system is completely automated and designed to work at huge scale. On the other hand Spence and Jo Bertram argue that surge prices brings more capacity into the market. In fact Uber claims there were twice as many drivers on the road in London on Thursday morning than normal.

But is the surge a force for perfect market efficiency? In the wake of a major snow storm in New York that saw Uber surge to 8.25x the New York Times hit the nail on the head.

But there is no way for customers to gauge supply and demand for themselves beyond looking at the dynamic-pricing multiple. And dynamic pricing is still not the same thing as true market pricing — like an auction system in which riders and drivers bid for one another’s services. Its opacity goes a long way to explaining the frustration it has generated

Back to the original question – it may or not be fair but is it gouging? According to wikipedia, most US anti gouging laws contain a three part test:

  • is there a public emergency?
  • is it a necessary good or service?
  • has the pricing exceeded legally set ceilings?

To answer the question one has to ask if the strike created a public emergency in the transportation system, if transportation should be deemed an essential service and if there are price ceilings or norms beyond which Uber prices should not be allowed to escalate. certainly its something that Transport for London and the Mayor of London should be considering closely.

Personally, I’d much prefer a higher average fare than fluctuating, dymanic prices. The base rate is too low for the market as is clearly evidenced in Jo Bertram’s article. Surge pricing does lead to customer dissatisfaction – I know because my driver ratings fall during surge periods. But the opacity of the system is a real problem. I’m not convinced the process isn’t being gamed cleverly. During this week I’ve noticed periods of no availability at Wimbledon and yet prices stayed static. At other times, I’ve seen surges come and go in waves that would have meant it unfeasible for meaningful movement of supply. London’s airports seem always to be exempt from surge prices for some reason. Like many others when I know the surge is coming and going I will just log out and wait for the surge to come back which it inevitably does. Consumers are doing the same dance from their side. Is it market efficient welfare inducing beahviour. I’m not so sure.