Barney Harford was hired late last year to help fix problems at Uber, the ride-hailing company. Instead, he has created new ones.

On a conference call this spring with colleagues, Mr. Harford, the company’s chief operating officer, critiqued a new ad that showed a mixed-race couple, said five people familiar with the conversation. He debated aloud how common the pairing was among the audiences that would see it. He also said he found parts of the ad’s early cut confusing, mixing up two black women in the video because they had similar hairstyles, said the people, who declined to be identified because they have signed nondisclosure agreements.

Though Mr. Harford later told colleagues that he regretted his phrasing, his comments struck many on the call as insensitive about race. They said it was part of a pattern by Mr. Harford in which he talked about women or minorities.

Aspie_Muslim_Economist_ posted:

As fun as it is to shit on Uber, they seem to have gotten their shit together in Southeast Asia, and I worry it's a sign of things to come.

They sold off their business in the region to local Uber-clone Grab in March, and somehow, despite assurances to regulators that service would only improve, the new monopoly is jacking up prices and extracting huge profits. Taxi services had been largely crowded out even before the merger, and attempts by new ride hailing services to enter have been stymied by the chicken-and-the-egg problem of network externalities--benefits a user of a platform provides to other users. A ride hailing service with few drivers and few users has unacceptable wait times, so the only way to grow to a critical mass where you can compete on even ground is to sink a huge amount of money into subsidizing drivers and users until the platform grows. After the struggle for profitability seen with Uber and other "subsidize to grow" platforms, no one wants to waste that kind of money just to be able to compete on price with the now dominant and mature Grab. Grab is quickly expanding their platform into a variety of other services, such as food delivery, electronic payments, money transfer, and retail. This Amazon/Alibaba style one-stop-shopping strategy will make them even harder to challenge.

The Grab deal was initially framed as a blow to Uber, but it was actually a huge win. Uber gained an almost 1/3rd stake in Grab in March, so it's traded 100% of it's previous losses for a third of Grab's monopoly profits. More to the point, Softbank had major stakes in both Uber and Grab prior to the merger (and owns an even larger share of Grab now via Uber's added stake), so this should be seen as Softbank consolidating their ride hailing portfolio and eliminating competition between their subsidiaries.

It's easy to mock venture capitalists for throwing good money after bad with network platforms like Uber, but the prize they see--transnational super-monopolies that control not just access to a particular good, but the means or medium by which a people access a large portion of necessary goods and services--would amply justify the temporary loss, and outcomes like this suggest that they're not just stupid.

I guess that's the point I wanted to highlight. From the left, it's tempting to mock tech bros as degenerate buffoons, and people like Musk and Kalanick make it easy, but behind these pathetic personalities there's lots of money and lots of intelligent (if not self-aware) people who are betting that platforms like Uber will give them so much political and economic power that they justify years and years of massive losses. You shouldn't underestimate how canny they are--and how dangerous any goal they'd shed so much blood for is--even if their figureheads act like clowns.

the general question i see about Uber is, can they sustain that sort of move through rent-seeking if they can't do it through razor-thin margins (if they can achieve those even)? for how long, and what happens when their time is up? and in this specific case i'm curious to know what you think

Chthonic_Goat_666 posted:

is there an article showing that theyre extracting huge profits? the taxi industry was never that profitable and jacking up the prices beyond what other transportation would cost will likely send people back to walking/public transport and also lead to the re-emergence of traditional taxi services.

tears posted:

im confused by that post since when it comes to uber i think we have mostly been considering the viability of "taxi firm but big and with an app" from a profitability point of view, since taxis are a luxury item that most people cant afford limiting any sort of monopoly price hike? like who gets a taxi? and if they have to diversify away from that into paypal but for SE asia to be profitable thaen doesnt that just prove the original point? Also im still not really sure how effective a monopoly on overpriced chauffeurs would be...unless you could crush the wage bill to keep ride prices down which would make your taxi drivers angry and prone to burning down your hq. i think i posted about this itt somewhere

e: and also as 666goat post i cant see any evidence that grab is making a profit? they seem to keep financials under wraps but i saw something about a 3 fold rise in losses 4 2016 compared to 2015, do you have some info about a reversal? overall profitability or per ride profitability also please.

At this point, my basis for saying that is theoretical rather than empirical. As tears mentioned, Grab has very opaque financials, and there's no publicly available data for the four month since the merger. However, prices have risen substantially, primarily through dynamic pricing with exorbitant rates at high demand periods. Given that there has not been a massive drop in usage and the cost of provision remains the same (driver's compensation has not risen--that can be checked via archives of their weekly payment structure posts), I would be very surprised if they're not turning a profit. I should have said "likely profits", though.

In terms of taxis being a luxury service, I think that's highly dependent on the locale--specifically on labor costs. For example, I do a lot of work with refugee populations in Malaysia, and they are often dependent on Grab for travel to work. In the core, wages are high(er), so labor constitutes a large portion of the cost of a taxi, and it's much cheaper to choose a mode of travel that doesn't require so much labor. In countries with much lower wages, fuel and depreciation on the vehicle--no cheaper than in the core--are far more important factors, and the benefit that taxis or ride hailing provide--more efficient utilization of the vehicle leading to lower costs per mile traveled--outweigh the labor cost. Also, since seat belt and occupancy regulations are rarely enforced, many riders can share the same car and the same fare. Additionally, public transit and pedestrian access is not well developed in all these areas. In Malaysia, cross-walks essentially don't exist and sidewalks are rare, so using public transit means dashing across 6 lane highways in between cars going 60 km/h.

Taxis are the most likely substitute for ride-hailing, but I don't think you can say they're equivalent services, at least not traditional taxis. Apps like Uber and Grab make hailing a ride much easier, and the ride comes much faster, which makes ride-hailing a viable general mode of travel, whereas traditional taxi services only really work if you always need pickup in extremely high traffic areas. Of course, taxis services can create their own apps and platforms, but taxi services have already largely been hollowed out by Grab and Uber, leading to the critical mass problem I mentioned before, and remaining taxi services actually depend on Grab. Grab started as GrabTaxi, which essentially outsourced the app side of the business (rider/driver matching, fair processing) for taxi companies. More accurately, it pulled individual drivers into the service to force the taxi companies onboard. Because of that, if you want to get a taxi without calling an operator and waiting 30 minutes you also go through Grab's platform.

Even so, there is clearly going to be some switching to taxis because of the jump in prices http://www.freemalaysiatoday.com/category/nation/2018/07/09/now-taxis-cheaper-than-grab/
, but so far it doesn't seem like Grab has lost a huge amount of ridership yet, and many of the taxi fares will still go through their app. Again they don't make numbers available, so I'm primarily relying on first-hand observation of travel choices and discussion on Grab driver forums like https://forum.lowyat.net/topic/4478891/+12140 . WRT the article about prices above, it's worth noting that prices for Grab aren't higher than taxis across the board, but only during peak demand because of dynamic pricing. In the long run, it may be Taxi drivers develop their own hailing platforms and squeeze out Grab--I hope that happens. I'm not convinced it will, though. MyCar, an attempt at exactly this, is floundering so far.

I do think that their rush to diversify points to the vulnerability of a freestanding ride-hailing app. Generally, with these sort of network platforms, it's often possible to enter against a dominant player while expending less on subsidies if you move slowly, targeting a small segment of the market and slowly expanding--say, operating in just one city at first. If Grab twiddled its thumbs it might slowly lose market share that way. My concern is that, when that sort of diversification is successful, as with Amazon or Google, it can entrench the firm in a position of huge political and market power. When you control multiple facets of a person's livelihood, you can use bundle pricing (cf. Amazon Prime) or any number of other strategies to gain an advantage over a competitor that only offers taxis.

More broadly, while I don't want to claim that Uber or Grab or even Softbank specifically will succeed in creating a hegemonic platform, I think recent developments suggest they're making more progress than one could have claimed, say, a year ago, and that in itself is a cause for concern. I fear that, while most of these ventures will fail horribly (and most have), the form of business that they represent and that is being so aggressively funded will succeed. To justify all those failures, the successes would have to be enormous. IT is giving ever greater advantages to these centralized, designed markets. We can see the beginnings of this with Amazon, but as it and other diversified platforms absorb more and more economic activity, I worry that we'll return to the logic of the company store--monopoly capital that goes beyond controlling access to a single good, but controls access to all goods simultaneously, as well as the information about them and the information about you. There are obvious antecedents to this in company stores themselves, as well as Walmart in rural areas, but combining this absolute market power with the ability to charge individual specific prices and to observe all consumer behavior can potentially give platforms even more control over our lives and ability to immiserate us.

Aspie_Muslim_Economist_ posted:

Chthonic_Goat_666 posted:

is there an article showing that theyre extracting huge profits? the taxi industry was never that profitable and jacking up the prices beyond what other transportation would cost will likely send people back to walking/public transport and also lead to the re-emergence of traditional taxi services.

tears posted:

im confused by that post since when it comes to uber i think we have mostly been considering the viability of "taxi firm but big and with an app" from a profitability point of view, since taxis are a luxury item that most people cant afford limiting any sort of monopoly price hike? like who gets a taxi? and if they have to diversify away from that into paypal but for SE asia to be profitable thaen doesnt that just prove the original point? Also im still not really sure how effective a monopoly on overpriced chauffeurs would be...unless you could crush the wage bill to keep ride prices down which would make your taxi drivers angry and prone to burning down your hq. i think i posted about this itt somewhere

e: and also as 666goat post i cant see any evidence that grab is making a profit? they seem to keep financials under wraps but i saw something about a 3 fold rise in losses 4 2016 compared to 2015, do you have some info about a reversal? overall profitability or per ride profitability also please.

At this point, my basis for saying that is theoretical rather than empirical. As tears mentioned, Grab has very opaque financials, and there's no publicly available data for the four month since the merger. However, prices have risen substantially, primarily through dynamic pricing with exorbitant rates at high demand periods. Given that there has not been a massive drop in usage and the cost of provision remains the same (driver's compensation has not risen--that can be checked via archives of their weekly payment structure posts), I would be very surprised if they're not turning a profit. I should have said "likely profits", though.

In terms of taxis being a luxury service, I think that's highly dependent on the locale--specifically on labor costs. For example, I do a lot of work with refugee populations in Malaysia, and they are often dependent on Grab for travel to work. In the core, wages are high(er), so labor constitutes a large portion of the cost of a taxi, and it's much cheaper to choose a mode of travel that doesn't require so much labor. In countries with much lower wages, fuel and depreciation on the vehicle--no cheaper than in the core--are far more important factors, and the benefit that taxis or ride hailing provide--more efficient utilization of the vehicle leading to lower costs per mile traveled--outweigh the labor cost. Also, since seat belt and occupancy regulations are rarely enforced, many riders can share the same car and the same fare. Additionally, public transit and pedestrian access is not well developed in all these areas. In Malaysia, cross-walks essentially don't exist and sidewalks are rare, so using public transit means dashing across 6 lane highways in between cars going 60 km/h.

Taxis are the most likely substitute for ride-hailing, but I don't think you can say they're equivalent services, at least not traditional taxis. Apps like Uber and Grab make hailing a ride much easier, and the ride comes much faster, which makes ride-hailing a viable general mode of travel, whereas traditional taxi services only really work if you always need pickup in extremely high traffic areas. Of course, taxis services can create their own apps and platforms, but taxi services have already largely been hollowed out by Grab and Uber, leading to the critical mass problem I mentioned before, and remaining taxi services actually depend on Grab. Grab started as GrabTaxi, which essentially outsourced the app side of the business (rider/driver matching, fair processing) for taxi companies. More accurately, it pulled individual drivers into the service to force the taxi companies onboard. Because of that, if you want to get a taxi without calling an operator and waiting 30 minutes you also go through Grab's platform.

Even so, there is clearly going to be some switching to taxis because of the jump in prices http://www.freemalaysiatoday.com/category/nation/2018/07/09/now-taxis-cheaper-than-grab/
, but so far it doesn't seem like Grab has lost a huge amount of ridership yet, and many of the taxi fares will still go through their app. Again they don't make numbers available, so I'm primarily relying on first-hand observation of travel choices and discussion on Grab driver forums like https://forum.lowyat.net/topic/4478891/+12140 . WRT the article about prices above, it's worth noting that prices for Grab aren't higher than taxis across the board, but only during peak demand because of dynamic pricing. In the long run, it may be Taxi drivers develop their own hailing platforms and squeeze out Grab--I hope that happens. I'm not convinced it will, though. MyCar, an attempt at exactly this, is floundering so far.

I do think that their rush to diversify points to the vulnerability of a freestanding ride-hailing app. Generally, with these sort of network platforms, it's often possible to enter against a dominant player while expending less on subsidies if you move slowly, targeting a small segment of the market and slowly expanding--say, operating in just one city at first. If Grab twiddled its thumbs it might slowly lose market share that way. My concern is that, when that sort of diversification is successful, as with Amazon or Google, it can entrench the firm in a position of huge political and market power. When you control multiple facets of a person's livelihood, you can use bundle pricing (cf. Amazon Prime) or any number of other strategies to gain an advantage over a competitor that only offers taxis.

More broadly, while I don't want to claim that Uber or Grab or even Softbank specifically will succeed in creating a hegemonic platform, I think recent developments suggest they're making more progress than one could have claimed, say, a year ago, and that in itself is a cause for concern. I fear that, while most of these ventures will fail horribly (and most have), the form of business that they represent and that is being so aggressively funded will succeed. To justify all those failures, the successes would have to be enormous. IT is giving ever greater advantages to these centralized, designed markets. We can see the beginnings of this with Amazon, but as it and other diversified platforms absorb more and more economic activity, I worry that we'll return to the logic of the company store--monopoly capital that goes beyond controlling access to a single good, but controls access to all goods simultaneously, as well as the information about them and the information about you. There are obvious antecedents to this in company stores themselves, as well as Walmart in rural areas, but combining this absolute market power with the ability to charge individual specific prices and to observe all consumer behavior can potentially give platforms even more control over our lives and ability to immiserate us.

thank you for the detailed reply. just in general i think this discussion won't really "get anywhere" until we have actual financial information and companies like Lyft and Grab are pretty tight-lipped about their financials unlike Uber. i'm pretty busy lately so i do encourage people to post updates here, i simply dont have enough time to read as much stuff as i did a year ago for this thread.

let me go over a few countervailing tendencies i think could restrict Grab from making a profit.

Apps like Uber and Grab make hailing a ride much easier, and the ride comes much faster, which makes ride-hailing a viable general mode of travel, whereas traditional taxi services only really work if you always need pickup in extremely high traffic areas.

the reason why rides come faster with ride share companies is that most of these companies are simply hiring too many workers for the amount of work that they're doing. they may not get a good overall wage every day, but these ridesharing companies have to pay them for down-time and snaking routes that taxi drivers simply don't do. for consumers this is very convenient of course, but that doesn't necessarily mean its a profitable strategy.

these ridesharing companies also usually have a high employee turnover rate, which means they have to spend big to incentivise new employees, it also means they have a lot of newbie drivers who don't know efficient routes and cannot maintain their car properly leading to higher repair costs.

the network effects of these platforms i think are pretty negligible. presumably people are just chasing the cheapest ride and if there's a cheaper ride elsewhere i'm not sure there's really a reason to stick to one particular company.

i think the diversification strategy is probably a winning one for these companies, muscle into retail or finance or whatever using their massive venture capital cash and then use profitable business elsewhere to subsidise the visible ridesharing part of their company as a loss-leader. whether they'll be able to do that long term is something that remains to be seen.

Edited by Chthonic_Goat_666 (July 21, 2018 07:09:02)

tears posted:

thanks for writing all that. It raises the interesting thing that we (I) have probably been overlooking about the different function of ride-hailing apps in cities with ineffective public transport/gridlocked road networks at peak times, compared to european countries which have good systems and are less prone to megagridlock. like say in the uk with a very well developed public transport system, a taxi/ride hail is generally considered as a pricy luxury alternative to the bus(/tube if u are in london), rather than as an alternative to your own personal vehicle, which many people taking ubers own also, while i can see that in the cities of south east asia being able to hire a person on a motorbike to take you somewhere occupies a completly different position to both sitting in major traffic on the bus and to buying and maintaining an expensive vehicle for you to idle in traffic inside. is it different un the USA because afaik public transport is bad?

Im also interested in the fuel aspect because of the flashpoint fuel prices cause (like cars posted about haiti in the main forum just now, which was major sparked by govt rise in fuel prices), if fuel prices go up then the proportion of fixed to variable capital rises right? But mostly this would just increese the cost of rides and price the poorest users out? driving them (heh) to use other methods of transport (buses, bicycles?)

Yeah, I think the viability and profitability of ride hailing platforms varies a lot from country to country and even from city to city. Malaysia has bad public transportation, a sprawling central city, and few bicyclists (though many scooter and motorbike riders) because of an extreme car culture caused by decades of petrol subsidies. You walk or ride at your own peril, and nothing is close by. Many countries in SEA have much stronger bike cultures, and Singapore, for example, is a single city with constant gridlock and excellent public transit. You might expect ride hailing to be more profitable in Singapore because of the finance-inflated per capita GDP, but I think these companies actually have less to offer in that kind of city. I think many of these companies think of different markets as experiments, and may not expect ride hailing to succeed everywhere, at least not as a freestanding platform.

Chthonic_Goat_666 posted:

thank you for the detailed reply. just in general i think this discussion won't really "get anywhere" until we have actual financial information and companies like Lyft and Grab are pretty tight-lipped about their financials unlike Uber. i'm pretty busy lately so i do encourage people to post updates here, i simply dont have enough time to read as much stuff as i did a year ago for this thread.

let me go over a few countervailing tendencies i think could restrict Grab from making a profit.


Apps like Uber and Grab make hailing a ride much easier, and the ride comes much faster, which makes ride-hailing a viable general mode of travel, whereas traditional taxi services only really work if you always need pickup in extremely high traffic areas.



the reason why rides come faster with ride share companies is that most of these companies are simply hiring too many workers for the amount of work that they're doing. they may not get a good overall wage every day, but these ridesharing companies have to pay them for down-time and snaking routes that taxi drivers simply don't do. for consumers this is very convenient of course, but that doesn't necessarily mean its a profitable strategy.

these ridesharing companies also usually have a high employee turnover rate, which means they have to spend big to incentivise new employees, it also means they have a lot of newbie drivers who don't know efficient routes and cannot maintain their car properly leading to higher repair costs.

the network effects of these platforms i think are pretty negligible. presumably people are just chasing the cheapest ride and if there's a cheaper ride elsewhere i'm not sure there's really a reason to stick to one particular company.

i think the diversification strategy is probably a winning one for these companies, muscle into retail or finance or whatever using their massive venture capital cash and then use profitable business elsewhere to subsidise the visible ridesharing part of their company as a loss-leader. whether they'll be able to do that long term is something that remains to be seen.

Edited by Chthonic_Goat_666 (yesterday 10:09:02)

Yeah, I think the next couple years will be very telling. I don't think speculative funding will last any longer than that, so we'll see something conclusive, one way or the other.

The other way to frame Grab's SEA merger is as a last chance for ride hailing. The business model for these ventures is
1) grow by subsidizing both sides of the market (as you mentioned, they hire drivers just to be available, and they also subsidize fares for riders),
2) remove rivals and create a monopoly, and
3) create a profitable tent-pole for a larger conglomerate platform.

Grab has now executed 1) and 2) in many parts of SEA (some countries, like Indonesia, have Go-Jek as well). Regulators are complaining now and threatening to impose taxi regulations on Grab, but this is about the best outcome they could have realistically hoped for. If they can't make it to 3), either because the monopoly is too vulnerable to new entry or other modes of transit hold prices below profitability, it'd suggest the business model just doesn't work, even under ideal conditions.

I do think the rider/driver clearinghouse mechanism has advantages over traditional taxi service, at least in some settings. There's not much empirical research on it so far, but this unpublished paper ( http://www.shapiromh.com/uploads/8/6/4/0/8640674/mshapiro_jmp.pdf ) suggests that Uber's clearinghouse is
1) useless in extreme density areas where physical cab hailing is feasible, but
2) more efficient than cabs in lower density areas,
which is what you'd expect a priori. The other major result is that much of Uber's value comes from skirting taxi regulations--no surprise. The paper doesn't give any estimated relationship between pickup travel distance and number of users, which is where the network effects would come from, but some basic numerical simulations (randomly drawing drivers becoming available and riders requesting rides on a 2d grid, then matching them to minimize average distance) suggests doubling the number of drivers and riders cuts pickup travel distance by 20-30%, depending on what assumptions you make about the search and matching process. I'd guess the real number is closer to 20%, but that still suggests significant advantages for a large platform over platforms or taxi services with much smaller fleets.

Again, cab services can in principle utilize the same sort of matching app, but Grab is hoping that their huge size advantage will make them unassailable when combined with the dependence of traditional taxis on their app. Time will tell.

I can't believe I missed this profound bit of Musk Thought in a previous Tesla earnings call pic.twitter.com/DXElz7DTgP

— Pinboard (@Pinboard) July 25, 2018