Delivery service price cap regulations

Ben emails me:

Could you please consider and comment on some of the unseen consequences of local price caps on restaurant delivery services?  (Politico article describing the phenomenon in SF, NYC, etc.)  A highly competitive market for such services exists between GrubHub, DoorDash, Uber Eats, etc.  Moreover, patrons can always pickup and restaurants can always hire their own drivers.  That dynamic market will keep prices down and improve service quality and value.  As reported 2 days ago, 5/13/2020, in the Wall Street Journal, “America is stuck at Home, but Food Delivery Companies Still Struggle to Profit.”  Yet many locals are considering regulating and limiting the prices that such delivery services can charge.

Here is a NYT article on the same phenomenon, claiming that some apps charge up to 40% of the restaurant’s take.

My first question is why the restaurants do not charge higher prices for customers using the app.  That might be illegal in some localities, but surely that is not the general answer to the question.  Rather the restaurants are afraid of losing customer good will  — “what!? I have to pay 30% more just because I bought it with my phone?”  [Plus the apps do not allow it, see the comments, though I do no think the apps could prevent restaurants from giving “extras” and thus lower prices to those who show up for service in the restaurant.]

In this setting, restaurants are losing potential revenue to avoid a reputational hit, and staying in business (rather than closing up) because they believe the value of their future reputational franchise is high.  In other words, in both channels the restaurants perceive the value of their future reputational franchise to be pretty high.

That is the good news, although you might wonder how it squares with the generally low returns to running a restaurant.  I suspect some restaurants simply know they are good and profitable because they are skilled, and the losers are overconfident and less well-informed.

One efficiency advantage of the apps is that they will put the unprofitable restaurants out of business more quickly.

The next question is whether some surplus from the profitable restaurants should, in the short run (and maybe in the longer run too?) be redistributed to the app company.

The apps should increase the demand for the food from the good restaurants (easier to order and arrange delivery), but lower the profit margin on selling more of that food.  If those ingredients and kitchen capacity otherwise would go to complete waste, overall that seems like an acceptable bargain.  Kitchens are kept active, which is an efficiency gain, even if some profit is redistributed to the app company.

In this scenario, you can think of the app as doing some of the selling, rather than the restaurant doing that selling, and reaping surplus from that effort.  In essence, the business of the restaurant has become more specialized, toward pure food production and away from selling, that latter service now being performed by the app company.

Restaurants that were great at selling in the first place might be worse off.  But it is far from obvious that these apps and their prices should be decreasing efficiency.  Some other restaurants might be worse off because it is harder for them to carve up or segment the market, but that change likely is efficiency-enhancing.

And if the apps do indeed speed the bankruptcy of the lesser restaurants (presumably what the critics have to believe), over the longer haul prices will indeed go up and the good restaurants will earn back some of what they lost up front.

On net, consumers will have better services, better marketing, pay higher prices, and have a better selection of restaurants.  That just doesn’t sound so terrible, or so necessitating government intervention to cap app prices.

Note that informed customers probably need the app least, so they are least likely to see its value, just as “critics” as a class, including restaurant critics, are also least likely to see the value of the app in marketing the restaurants.  Of course this class of “critics” are exactly those who are most likely to be writing about the apps.


Makes currency useless. I've stocked up on cigarettes, brandy and nylon stockings.

"My first question is why the restaurants do not charge higher prices for customers using the app."

It doesn't work that way. When restaurants sign on to one of these delivery apps, they are working for the app, not the other way around. Restaurants will be told if they don't like the fees associated with an app, they are free to leave and develop their own instead.

Yup. The food delivery companies are using their bargaining power to force the same price for delivery and eating in. This is similar to credit card companies that force retailers to charge the same price for those who purchase with card or cash. Both scenarios have a distortionary subsidy effect since the cost clearly isn't the same.

"A group of New Yorkers sued GrubHub Inc., DoorDash Inc., Postmates Inc. and Uber Eats, accusing them of using their market power to stop restaurants from discounting meals to customers who order directly, even though it’s cheaper to do that."

A competitive marketplace exists, so if all apps are doing it, then that seems to be an indicator that price discrimination by service type (delivery/dine in) isn't very efficient, right?

> That’s forced companies like GrubHub to offer food discounts and reduce or eliminate delivery fees. GrubHub on Monday cut earnings guidance as demand suffered in New York and elsewhere.

> The New York customers, who seek class-action status, say the delivery services charge “exorbitant fees” that range from 13% to 40% of revenue, while the average restaurant’s profit ranges from 3% to 9% of revenue, making delivery meals more expensive for eateries.

If grubhub is simultaneously "exploiting" it's "market power" and not making much of a profit... who is wrongly winning at the expense of whom?

Also, looking at the GrubHub T&C and FAQs on their website, they don't require in-store prices to be the same as take-out prices, and the only money they take is a percentage 'marketing fee' that you can modify based on how much you want them to advantage your business in the search results.

Uber itself routinely changes menu prices and charges the purchaser higher menu prices. I don't think its well-known that restaurants are unable to price discriminate. They may already be doing it. One time I went to popeyes to get there chicken sandwich. In restaurant it was $3. On the app it was $5. I'm not sure whose changing the price.

I work in the restaurant business. Restaurants have the power on most platforms to change the pricing these days. It's generally 15% higher pricing on the apps, so that they can pay the app fees that the restaurants pay the app (15-30% of order price). App gets paid from both the buyer and seller.

It's like the credit card agreements that prohibited either adding a surcharge for credit cards and a discount for cash, especially with the fee varies by merchant.

Congress has made those terms moot, but 98% of merchants just added the cost to all prices, but refuse to accept Amex, Diners, etc which have the highest fees. A few places offer discounts for cash, like my dentist.

Now that I think about it, it is the same story not just with credit cards but also with Amazon and iTunes. In both cases, tech companies entered into a new sector of the economy and quickly acquired enough bargaining power through the user base and data that they controlled to push down prices.

From a certain narrow perspective, this is a good thing because it lowers prices for consumers. But if we take into account the long-term effects of consolidation in restaurants, publishing and music respectively, it is much more dubious. There is a potential loss of variety for consumers and loss of bargaining power for workers.

"loss of bargaining power for workers"

That depends on the worker. Those who write the API* will make out big but those who are take orders from the API will be crushed.

* API = application programming interface

Yes. As people in my profession (antitrust enforcement) call it, by analogy of international trade law: the apps use narrow MFN clauses. (They'd use wide MFNs too, but those are illegal in many jurisdictions, because they are so wildly anticompetitive.)

I like Ricardo's explanation. And for what it's worth, I think it can be generalized. Certain technological waves (credit cards, magnetic stripe credit cards, digital payments so far) have created opportunities for intermediation. As opposed to disintermediation.

Sure, a store or restaurant could implement their own solution, but as everyone on this page recognizes, that brings zero network effects (nor economies of scale). The service is, more than the function, about the scale.

Could GrubHub be disintermediated? Sure, but it would face the same problems as disintermediating Facebook or Airbnb. The name has become the destination, providing the centralization and the intermediation.

This seems stable for now. Wild eyed technologists have always pictured alternatives, from 1980's mesh radio dreams to 2010s blockchain. But none of them have been more attractive than .. just going to Amazon to buy a mouse trap.

Restaurants don't have to sign up for these apps for the delivery services to advertise them as available for delivery. Since the delivery services are better capitalized, they can afford good SEO operations. Even before COVID, if you searched for a restaurant by name and city, its own web site would appear miles down below various aggregators offering reservations, travel arrangements, alternatives or delivery. Unlike the old Yellow Pages, search engines can be gamed. If a restaurant wants to handle its own deliveries, it will have to ransom its name from the SEOs.

To your first question - I believe that the contract involved in signing up for a delivery app usually restricts differential pricing. The platforms will not let them on if they're charging more to people who come through the platform.

Long established practice and thoroughly common.

Yes, it has been that way. But other means of payment are charged less. Commonly in gas stations now. With all our digitization, such a practice can spread much more quickly.

The tactics of some of these apps to create growth are more than questionable, if not outright fraud. For example, a common marketing trick is to hijack the google maps entry for a particular restaurant so people looking up the restaurant's phone won't call the restaurant directly but instead order from a delivery service that takes a 30% cut or so. If you behave that unethically, you deserve more regulation.

Here's the relevant hacker news discussion:

Yup, and also creating fake websites, hoping the customers will order from that website thinking that they're ordering from the real restaurant.

Free market competition, ain't it grand?

Free market competition is indeed grand. Surely that's what's established by analysing the anti-free market behaviour of the high tech Robber Barons.

That's fraud - not the free market

"but instead order from a delivery service that takes a 30% cut or so."

Why would the restaurant give them a 30% cut in that case? If there's no agreement, there's no reason too. If there is an agreement then the restaurant isn't being scammed by a high jacked google entry, it's something they agreed too in the first place.

It's a dispute on the agreement. If the app opens a new channel through it's branded website then it should get a cut from the restaurant by sending them customers from their branded website. But if was already made to take orders and the app creates to redirect that business through their app from people who otherwise would have gone to .com that's not adding customers which is what the restaurant is paying for.

True, but I believe that's actionable. The restaurant can just demand that the competing url be shutdown or handed over unless they signed an agreement giving the app rights to that specific url.

Just to be clear, if companies are illegally doing this kind of action, then they should clearly be punished.

Will there ever be an example where cutting out the middle man is rewarded? - "The next question is whether some surplus from the profitable restaurants should, in the short run (and maybe in the longer run too?) be redistributed to the app company."

And isn't the term redistribution seen with extreme disfavor among those TC has lunch with?

"Will there ever be an example where cutting out the middle man is rewarded? "

You can check out the history of the Soviet Union. They were big on cutting out the "useless" middleman.

The Party quickly became the universal middleman, and since it was the very definition of useless, the result was plain.

To be fair, that's not really an accurate explanation. The USSR did actually greatly reduce the numbers of middle men in order to increase efficiency in the supply chain. Unfortunately they found out that the middle man often added valued as a source of lower level decision making, stockpiling, expertise and feedback . It turned out that the middle man often added more value than his/her costs.

The problem with the USSR was that it was just a rebranding of Russia.

These delivery apps have increased the prices for delivery 20% or more for little added value because of their Oligopoly. Where is the value added?

I'm ordering the same things I was order 3 years ago from the same restaurants, my costs have gone up, and the restaurants net income goes down.

>Where is the value added?

In ordinary times, it probably let the restaurant sell a few additional meals every day. That is, I'm assuming table space is the key constraint, not kitchen capacity, in most restaurants.

In today's environment? Not much choice. OTOH, the restaurant isn't paying serving staff, so maybe it's a wash from their point of view.

I think you’re misinterpreting unprofitability for a restaurant to mean poor quality product (food) when in reality it’s often just the opposite. Chain restaurants are highly profitable in large part because their product is low quality.

Looks like one of Tyler's trolling exercises. Notice that he talks a lot about efficiency and demand, but edited out any opinions he might have about quality.

"In this scenario, you can think of the app as doing some of the selling, rather than the restaurant doing that selling, and reaping surplus from that effort. In essence, the business of the restaurant has become more specialized, toward pure food production and away from selling, that latter service now being performed by the app company."

In a customer facing business, you never let someone else get in between you and the customer or you become a commodity. I can tell you economists have never run a successful business before.


I really love that line and not sure I heard it. Also not sure it needs to be restricted to customer facing.

"you never let someone else get in between you and the customer or you become a commodity."

Check out Ben Thompson's Stratechery web site for more on this. It's one of the reasons Steve Jobs opened all those Apple Stores.

+1, the best response. Though, it's always true in advertising that there is a middle man, but usually the customer (the restaurant chain) has substantial control of the output.

One thing that isn't addressed here is how effective delivery applications are going to change the industry. With the rise of delivery apps, I could see restaurants focusing more and more on delivery only, thereby saving money through lower rents and payroll. This would also address the problem of differential pricing, since all food would be delivery or take out.

if I were in the restaurant business I would be strongly tempted to open a store that only has cooks and one or two expeditors. You could quickly produce relatively high quality food as they're being ordered, and you wouldn't even have to wait for customers to finish eating to start preparing the next order. Volume can go way up due to the delivery apps, and you can be left to focus on quality control.

The trick is in getting somebody to eat it once the COVID-19 threat has faded. Warmed-up food doesn't taste very good, people go to restaurants for the social life and to be waited on, and if you want prepared food you can get it at good grocery stores. Yours is a niche business idea that has already been tried and is not very successful.

Of course I only speak for myself, but going to the restaurant is generally the least favorite part of the eating out experience for me. The kind of restaurants I attend, which are not high-end since I can't afford it, are generally crowded, noisy, frequently difficult to get timely service, and are usually incentivizing the patrons to eat and leave as quickly as possible. I've worked at such restaurants.

On the other hand, having friends over at your own house and getting food delivered allows fantastic food, better than I could prepare myself, without the hassle of making it or the hassle of cleaning up. I don't even have the hassle of having to go and get it, just order it and it comes in trays designed to keep the food hot, we all eat and it greatly enhances the dining experience for me.

As far as the quality of food, the main reason I enjoy going out is it allows me to eat food I otherwise wouldn't eat. When I get delivery or take out, I find I enjoy it more and can't pay more attention to this aspect of eating which the restaurant experience actually detracts from.

Larry, warmed-up food does taste good, if the warmed-up food is selected with some care by the restaurant creating the menu. I get that some foods are not going to work as carryout. For instance, anything like onion rings that tastes best hot and crispy is going to be a no good by the time it gets to your house 15-20 minutes after it leaves the kitchen. So restaurants that focus on delivery will make foods that survive or even benefit from the delay between leaving the kitchen and arriving at your door. Think stews or sandwich-kits which you assemble yourself. Salads with dressing on the side, etc... Chinese restaurants in our area are doing well. As are pizza places.

Why not allow people to prepare food for delivery at home? Let's not make it illegal like they did to a woman who sold food on Facebook.

This already happens, except in a stronger form that you are describing. They are called ghost restaurants: they have a name and a menu, but there may be a half dozen of them operated by a single company that rents a couple of big food trucks and all of the orders for all of the restaurants are cooked in those trucks and then delivered.

"if I were in the restaurant business I would be strongly tempted to open a store that only has cooks and one or two expeditors."

That's what our local Chinese restaurant has been doing for ages. We can collect or they can deliver. Surely this must be very common?

+1, he's just describing the standard Take Out/Delivery only restaurant structure.

The trend in China is towards "ghost" restaurants that share similarities to the traditional American Take Out/Delivery restaurant but have some important differences. The difference between the Chinese-as-in-China model and the Chinese-as-in-General Tso's model is that in China the restaurants will be "headless" and at your local Chinese restaurant you can still walk in the storefront and order at the counter. The innovation that has taken place in China is that new restaurants can rent kitchens like that at a culinary school and can only be ordered from with a delivery app like Meituan. They save on the costs of a storefront such as decorating and securing a street side store front, handling deliveries in a back alley, decorating the interior and creating menus for the Luddites who enter as foot traffic, hiring a dedicated cashier to take orders and handle cash. These kinds of innovations are what allow certain new categories of restaurants to survive in the extremely competitive restaurant market.

That would strike me as a niche model, designed for high density cities that already have a large restaurant capacity.Traditionally most Take Out /Delivery places have a higher margin on their Take Out even when factoring in the additional costs.

The front person is taking phone and/or computer orders in any case. The store front is minimal and acts as advertisement and to get walk/drive in traffic. And of course, maintaining delivery staff is expensive.

You're right, we shouldn't be surprised to see this type of restaurant in the country with 8 megacities first.

What's missing in your analysis is the coronavirus, the rise of cooking at home, the realization among many that real estate is hugely overrated in the internet age, that app companies are unprofitable burning through VC cash without a business model that may end up being dubious (anybody remember delivery tech cos like Webvan or Cosmo?), the impact of small business once again eaten up by big business discouraging new entrepreneurship and business formation.

"that app companies are unprofitable burning through VC cash"

That's just the usual tech business model, though. Companies plan on losing money through their first few years of operation with the goal of acquiring a foothold in the market. Then, they raise their prices and focus on consolidating their gains to start turning a profit. It is a winner-take-all game in which most apps will go the way of Friendster and Myspace but the ones left standing will probably become giants.

As to why the business model makes sense, plenty of pizza and wing shops have offered delivery for decades. In the developing world, even McDonald's has offered delivery for years. It is not difficult to make earn revenue or find consumers willing to pay for food delivery -- the problem has always been the logistics. Running a restaurant is already a huge amount of mostly thankless work without also having to manage a pool of delivery drivers, resolve disputes over payment, late deliveries, cold food, etc., and maintain a website with an updated menu so people can call in an order. Outsourcing these problems to another company is an attractive proposition.

It isn't going to work, though. Once Softbank and the other VCs get bored with subsidizing Grubhub's and Uber's customers, they are doomed. When they're already eating 30+% of restaurants' take, they will have to go to customers for more, and there is no way in hell people are going to start paying $10+ more for a pie to have it delivered. There is zero reason for customer loyalty (maybe Uber can cobble together a meals & wheels service that would appeal to people like their engineers...).

Fact is, there is no economically significant economy of scale here take advantage of, and when you stick a software/logistics multinational in between a local restaurant and a local hungry person, the restaurant makes less and the hungry person pays more for worse service.

Hard to understand so much speculation on an issue conditioned by "America is stuck at home..." (WSJ) or "But once the lockdowns began..." (NYT). The relevant issue is how lockdowns have affected business, in particular small business owners. Too many have been forced to close as reported in

Please pay attention to the much larger effect on minority businesses.

And survivors have been forced to adjust in ways that increase the cost of delivering to consumers. How is the additional cost distributed between producers and consumers? Just rely on Econ 101 for a simple anwser or on IO for a sophisticated one. Anyway, pay attention to expectations about how long consumers will be stuck at home.

"My first question is why the restaurants do not charge higher prices for customers using the app. "

The apps don't allow it. It's part of the terms of service.

Others have already pointed out that the app companies have illegally hijacked restaurants websites to make it hard for them to do their own delivery.

The impression I get from looking into this is that the apps are very inefficient in organizing delivery and use the huge fees from restaurants to run big unsustainable promotions to try and draw customers, but like all unicorns they just care about top line growth even if it's unprofitable. Their exit strategy is an IPO, and if investors end up holding the bag who cares. It's like WeWork. There is no IPO for restaurants, no greater fool. They have to make money.

We are all Amazon now. By that I mean the Amazon retail business model: the product comes to the customer rather than the customer to the product. Of course, it makes no sense in terms of efficiency but it worked, and worked like a charm, for Amazon. But only because Amazon could operate at a loss, a loss funded by a rising stock price. Until Amazon found a business that it could operate at an enormous profit to offset the losses in its core business, a business that is only tangentially related to retail: the cloud. If Amazon's path is the path of success for retail, then Amazon's path is the likely path for restaurants (and grocery stores too). In the case of restaurants, that means the end time for the locally owned specialty restaurant. For economists who believe big is beautiful, that's not a bad development, but for economists who believe in many small firms operating on the margin in a competitive market, it really is the end time (for capitalism as envisioned by the prophet Smith).

I ordered delivery from a restaurant last night. I was charged a $5 delivery fee and a $1.50 "convenience fee." I assume one of these fees pertains to the app and the other pertains to the restaurant, but it's possible that both came from the app. Either way, I'm paying higher prices to get my food, so a lot of the commentary above is just a head-scratcher for me. The customer is paying more for the same food. Somebody has to cover the cost of the delivery, and that somebody is the end customer. The app company and the restaurant can fight over which one gets what percentage of my $6.50, but from my vantage point, everything is working out in a fair and expected way. What am I missing?

Let's say you order a $50 dinner. You pay the $6.50 which goes to the driver. What goes to UBER is $50 * 40% = $20. The restaurant only gets $30.

Why can't UBER make money on the $20. Well, it runs these promotions like 30% off your next five orders. Or it pays driver incentives. Or it pays for advertising. You know, that wonderful stuff unicorns do burning through VC money to acquire customers.

The restaurant doesn't want to run those promotions and get into a zero sum bidding war where revenue is below their costs. Unlike UBER, they aren't hoping to exit via IPO. They actually need to run profitable businesses.

"The restaurant only gets $30."

(a quick google search indicates that Uber gets 30%).

I'm trying to see your point here. It's a voluntary transaction and the restaurant can always refuse to do business if they think the cost is too high. Honestly how is this different from the restaurant offering a 30% coupon on low volume nights? The restaurant is trying to generate additional marginal business by lower its costs in both cases.

30% may be right and matches my memory on UBER, though the post leads with "claiming that some apps charge up to 40% of the restaurant’s take" and that makes for easy example math.

"It's a voluntary transaction and the restaurant can always refuse to do business if they think the cost is too high."

The alternative up until now has been running a restaurant, but that was made illegal. Even running your own delivery has been subject to slimy practices (apps stealing their clicks online).

I'm unaware of any apps with fundamentally different terms. I think a big part of the problem is that none of these delivery apps are looking to run a profitable business, so any app that didn't run at a loss would probably lose out to one that would. But restaurants aren't in the kind of business where they can run a loss and burn through VC money.

"Honestly how is this different from the restaurant offering a 30% coupon on low volume nights?"

Because the UBER promotions are all day, even at peak times. They eat into what should be the high margin business of the day.

Fair point, but I still don't see that this is a big deal. A restaurant can just not sign up with any of these App services. If it's truly a bad deal, they say No.

Thanks, that was a helpful explanation.

I didn’t think there was any big mystery on the pricing question. It’s a little surprising, because when I was a kid it wasn’t unusual for local pizza joints to charge $2 or $3 for delivery.

I’ve been watching restaurants along Wisconsin Ave/Rockville Pike in Montgomery county over the past couple of months and it’s been interesting. There is a mix of standalone, local chain and national restaurants.

A lot of places started pickup/takeout as soon as allowed and were only fully shut down for a week or ten days. Some have yet to reopen at all even though outdoor seating is now allowed. There hasn’t been an identifiable correlation between ownership or food type and reopening status.

It does seem like several places that were barely viable before the pandemic have closed permanently. When, for example, one bagel shop on the main drag closes and goes out of business and another one 3/4 of a mile down the road starts doing takeout counter service ASAP and is still in business it seems a big assumption to blame either the virus or government’s reaction to it for any particular business failure.

On the delivery issue - I’ve found the best thing is to walk to a local place, order and pay, then pick it up in 15 minutes. Often I find an extra order of wings or something like that when I get home. It must cost them less to incentivize people ordering directly with a bonus appetizer than it does to pay the delivery fees, and it’s a way to get around the pricing requirements in their contracts.

Doordash, Grubhub and any other business that thinks they will get 30% of the ticket for taking the order and delivering the food will be out of business when they run out of VC money.

I'm not a restaurant owner, but I live next door to one. She informs me there are very few restaurants in the world that can give up 30% of their gross and make a profit. The other thing she said is local restaurants are using Doordash to expand their takeout business, then when it looks like it's going to work, they bring delivery in-house where they can attempt to control the costs and pass the delivery costs on to the delivery customers.

Restaurants that rely on alcohol sales to make a profit, like most "fine dining" establishments, are not going to do well at all in the delivery business.

"then when it looks like it's going to work, they bring delivery in-house where they can attempt to control the costs and pass the delivery costs on to the delivery customers."

That seems like the logical reaction. They are using Doordash, Grubhub as expensive upfront advertising to expand into a new market. Once they're established in the market, they'll be free to find a lower cost distribution only service.

The prices charged by the grocery store for items purchased via a grocery store delivery app (Instacart is the one I am familiar with) are in some cases higher, so the purchaser pays both the app (I pay a recurring monthly fee) and a higher price for the groceries. The app also itemizes my "savings" by purchasing via the app but does not itemize the higher prices charged by the store. Transparent the arrangement is not. It's a convenience, so one should expect higher prices and opaque pricing. Just like Amazon.

" a better selection of restaurants" I'm not sure how this works. Are you saying that by eliminating the restaurants of poor quality, it'll help increase the number of high quality restaurants by removing the competition. The end result of which would be that although there are less total restaurants for consumers to choose from, the selection that is available will feature more high quality ones, so overall it's a greater good?

Personally, my alliegance to restaurants is based on one-part food choice & quality and one-part atmosphere/intangible.

Sure, there's places in strip malls and nasty servers with amazing food. But mostly we go to places in part because we like the view, or the decor, and the vibe we get from the staff.

During covid, I support take-out from the places I hope to one day dine in again at. And my enjoyment of the meal includes the obvious taste characteristic, but also intangible and memory-based rewards, along with anticipation for the future experiences. I cannot separate my emotional connection with the restaurant from the demand to order out.

In other words, the take-out demand is neither based solely on quality, or merely price-driven.

So, in my view, the ordering apps seem likely to bifurcate demand into two separate business models: dine-in restaurants who build loyalty around the dine-in experience, and strictly take-out players, who maximize quality at the necessary price points, and do not bother with physical presence. It does not seem economically possible to do both.

And what existing restaurant owners sense, or ought to, is that the worst thing they could do is cannibalize loyalty to their physical version in pursuit of online sales. They need to do one, or the other.

There is nothing more annoying the visiting a fast casual restaurant, and being forced to wait and watch while the staff takes a phone order and puts it ahead of you in the queue. More than once I have been tempted to walk out to the sidewalk and phone my order in.

It's not like there's some equitable distribution of revenue that's being distorted. The basic problem is that delivery is not profitable for anyone. The apps can't make a profit even at rates where the parasite is killing the host. Before the apps came along, outside of a few dense urban areas, delivery was common only in a few specialized niches: Chinese restaurants (where some family member does the driving) and pizza places (which are often just kitchens with little wait staff). Customers know those foods travel well, so those places expect enough delivery business to invest in their own infrastructure. Like JimG says upthread, other places nowadays are hoping to grow to the point where they can justify their own delivery staff, but there's not much room in the market for that, and no reason to think the apps are going to help them out-compete other restaurants.

It's not clear what the stable equilibrium eventually will be. California is trying to kill these apps via regulation, but it seems like sooner or later the market will do so anyway almost everywhere. Perhaps drones will be the future here.

The restaurant business is officially the worst business of 2020. If you somehow survived the COVID lockdowns, the botched SBA loans, and all the looting and burning, you have giant tech companies with unlimited money looking to not only destroy your margin but put you into the red and lie about it. People wonder why small business formation is so bad these days.

Separate into two businesses - "Food Prep" and "Place to Eat". A $10 (tip licluded) meal at a restaurant is $6 of Food Prep and $4 Place to Eat. "Home Deliver" could end up being more or less than $4 when the dust settles.

A friend of mine, a child of refuges from mainland China, visited Hong Kong with his family back in the 1970s. The custom at one market was to choose your fish from a fishmonger who had tanks of them, then arrange to have it prepared by a cook who also provided a place to eat. His mother chose a fish. She then told the cook how to prepare it. The fishmonger said, "waste of a perfectly good fish" and tossed it back in the tank.

The "same price as eating in" is something all those delivery companies will absolutely force on the restaurant, because they know that otherwise, customers would be able to discriminate on delivery company pricing too: Then we'd have a VC fueled race to the bottom, and all the companies fail: The costs for the delivery companies are more than high enough to make them unprofitable, even when they rarely are a good deal to the restaurant, and once you add tips and delivery fees, they are still expensive for the customer outside of promos. It's not that they specialize in delivering the food on time and in good condition either.

So when you put it all together, food delivery like this is far worse value than all the companies that tried to just deliver from dark kitchens, but those don't do well either.

To give us a useful jump in food delivery we need far higher fixed costs than anyone wants to pay: Company provided equipment to keep the food in great shape, short delivery times, and recipes designed to handle the delivery time and be good for reheating: More like a pizza delivery business than "let any restaurant delivery any dish anywhere"

There are many options for this:

1. Joint ventures among restaurants to integrate into delivery service market. Could be done through a restaurant trade association.

2. Joint venture with restaurants through a credit card provider for it to negotiate delivery service and coordinate delivery. You probably already have a credit card app on your phone.

3. Scheduled delivery service: Before 3pm you schedule time slot when you want your meal and restaurant arranges delivery service, introducing competitive bidding.

4. Invent a new service: concierge meals: Andrew Zimmer or some food personality surprises you nightly with a meal they choose, based on your food allergies, flavor preferences, etc. Website tracks your reaction and review.

4. Sous vide food delivery service: you finish off the meal and claim it was your own.

5. Work with ethnic mom and pop restaurants to create a faux catering service that displays the diversity of the community and serves the food out of a food truck. You track the food truck and appear to collect the food when it is in your neighborhood. Any paid for and uncollected food gets delivered to the last stop: a local homeless shelter.

If the pandemic goes on for a while, we'll probably see more innovations like this, and there's a good chance many of them will stick around even after the plague dies down.

What I am seeing in Seattle: pizza places have their own ordering apps/web sites. One pizza place directs you to order on the website instead of call. Makes sense, once it is paid for, a website is cheaper than someone to answer the phone and transcribe your credit card number and other info.

Also, a large part is that restaurant delivery just doesn't work so well except a few select niches. If it were, it would have been prevalent already before the apps. Instead, it was pizza (everywhere) and maybe some Asian restaurants depending on the neighborhood. Probably depending mostly on density.

A delivery driver for a night is mostly a fixed cost. In theory the delivery companies make most of the cost variable instead by making delivery fungible. The cost doesn't go away. So how much will the market bear for turning the fixed cost variable? That is the profit opportunity for the apps.

Universal delivery will go away once the VC money dries up. Maybe pooling delivery will cause more neighborhoods to have more delivery options, but not everywhere like now. More restaurants will outsource tapping info onto a screen to their customers.

"Also, a large part is that restaurant delivery just doesn't work so well except a few select niches. If it were, it would have been prevalent already before the apps."

I think the main challenge has always been logistics. You need to hire and manage a delivery team, deal with payment procedure (some idiots think it is a funny "joke" to order food using someone else's name and address -- what to do?), dispute resolution mechanism, and a way of publishing your menu. Yes, density -- especially proximity to a college campus -- also helps.

The attraction of the delivery apps is that restaurant managers get to outsource all of the headache to another company. You always get paid electronically for the order once it is placed, someone else is managing the deliver drivers, and I imagine updating the menu displayed in the app is a simple procedure, so no need to maintain a website or print flyers.

The downside is that these services poach customers who would otherwise just walk in and order take-out and probably even a fraction of dine-in customers as well.

A walk around Manhattan and one will see posted in window of most restaurants is notice that orders can be made online: Seamless, etc. So, i guess they ok with charges.
No discussion of the pay to the delivery people and implications on cap and what means for their pay.

In NYC they limited the marketing fee to 5% (from 20%); however, they kept the 15% delivery fee (for those restaurants who do not use grubhub drivers) specifically so their pay would not suffer

The thing to remember is that many of these "disruptive" tech plays aren't required to make money as a business. Talk about an uneven playing field.

They are just required to generate enough funding churn to make money for the investors, leading ideally to an IPO - where the magic of the markets turn a money-losing bad business model into a unicorn.

The whole model is built around finding a sector where the incumbents are chained to actually having to make money, and then stealing their customers as fast as you can until the walls cave in.

Interesting issues on two sided markets:

1. In order to attract drivers, you need dense markets so that they are not waiting for an order and can deliver efficiently in a compact network.

2. In order to get a dense network, you need a number of participating restaurants to make you want to go to the app over another app.

Which came first, the chicken or the egg.

I now get drone delivery. Superman is on strike so he is no longer available to serve and leap over tall buildings in a single bound.

Also, there is a market for commercial kitchens and offsite prep. Expect to see branded commercial kitchens delivering recipe created food from chefs around the world.

If you get accustomed to take out, why not step it up a notch.

Not sure there's enough margin to go around for this otherwise nice idea. This market currently exists from the generous subsidies billions of VC dollars. Buying a $1 in revenue for $2 in costs is currently how Uber/Lyft and DoorDash/GrubHub work in the short term but what exactly is the long term vision?

Have a friend who owns and rents out commercial kitchen space. His business has not been affected as much as restaurants. Some of the customer include grocery stores, which remained open during the shutdown.

The delivery service part of it is an issue as you identified. If you look below I discussed the two sided market and density issue.

Those firms that fail, with VC subsidies, get acquired by those who are patient enough to wait. The issue, though, is density of network on the supply side and on the demand side.

As for buying revenue at a loss, I assume they are thinking that people will be forming habits, and plan to reap from the addiction to an app later. We'll see.

Glenn, good comment.

Here is a NYT article on the same phenomenon, claiming that some apps charge up to 40% of the restaurant’s take

Isn't the fact here the restaurant's kitchen can pump out a lot of meals but only X number of tables exist at the restaurant for people to sit down and eat?

Hate to say it but 40% may not be unreasonable. How much are you paying for the table to sit and eat (which probably takes up more than 40% of the space the restaurant is renting after all) versus the actual ingredients of the food and its cooking? If a restaurant with only ten tables can do 40 dinners without take out (two people per table, let's say two 'shifts' of dinner) but with takeout is able to sell 120 dinners then the cut could be quite reasonable.

People sitting in the restaurant order overpriced drinks and overpriced booze. It pays for the physical space. When they dine at home they buy wholesale liquor for 1/4 the price.

Perhaps but the question for the restaurant is only the capacity of the kitchen and does sending a dinner out the door at 60% of the menu price cover that cost or not. People who order out are by definition not going to the bar anyway.

The question then becomes does offering take out cause some tables to be empty? Then maybe not allowing take out may make sense to force the tables full. Or the restaurant could simply rent a smaller space with fewer tables but have a kitchen with more capacity.

"Hate to say it but 40% may not be unreasonable. "

Why don't you look up the actual margins of running a small restaurant before deciding what is or isn't reasonable? You might learn something.

As I stated before, the overall margins have nothing to do with it. When being confronted with a take out order, the restaurant has to ask what is the marginal cost of the kitchen pumping out one more dinner? Most likely the costs are very low. The kitchen in most places probably is capable of serving more tables than there are in the front of the house. Adding more tables to the front, though, is very costly since rents are usually pretty high. Most places end up with more kitchen capacity than tables. Hence the reason many try to push catering, since that is a place where you can put some of the excess kitchen capacity to use.

If your restaurant is going to survive, it probably matches kitchen capacity and seating capacity during peak periods, possibly lunch, possibly weekend evenings. You have to overcapitalize to some extent and balance costs by adjusting staffing. Delivery apps will use some off peak capacity, but they'll bring you down at peak hours unless you restructure your business.

I doubt this is possible. Kitchens are very efficient things if you know how to set them up. I'm trying to think about the cafeteria in my high school and the kitchen behind it. I don't think it was bigger than the kitchen you'd find in a Panera, could be wrong.

Long story short, tables are very inefficient with big space between them and comfortable seating. Kitchens are very tight and efficient. You probably can't just make your kitchen small enough to match your tables because commercial ovens, stations, etc. have min. sizes and a good kitchen staff makes efficient use of them. Any restaurant that sizes to their tables probably has a kitchen that can turn out more food than tables unless you're talking some type of super high end place where each dish is the personal creation of some type of celebrity which case you wouldn't do 'take out' anyway.

Grubhub generated $183M in cash flow from operations last year.  They generate a lot of money by charging 20% for being an intermediary.  Not sure how much money they make on the 10%+ they charge for providing delivery people to those restaurants that want this service.  I suspect they may lose money on this service.

The genius of their business model is if there is a problem with a client (make no mistake - the client is the diner, not the restaurant) they are quick to comp the entire order.  Sure Grubhub is out its cut of the order, but builds excellent brand loyalty.  Meanwhile the low margin restaurant, gets zip.  What is particularly infuriating is that, in my experience,  most complaints from diners actually relate to the incompetence of GrubHub's own delivery people, so effectively they are using restaurants money to compensate for their ownshortcomings.  Brilliant.

Some nights literally we could have 20% of our sales "refunded".  Literally a $100 order could be missing a $5 item and the whole thing would be comped.  Turns out their customer service is so woefully understaffed and poorly trained the reps would just automatically comp everything to get through the volume of calls they had.  Then, and here is the real genius, it takes 20 minutes to get to customer care to challenge a refund.  Then wait to be connected to restaurant care (no you cannot call them directly) and spend another 20 minutes going through what happened with them in the hope of getting the refund reversed.

The 20% GrubHub charges as an intermediary will attract more competition and get greatly reduced.  In the meantime a lot of restaurants will suffer...

Canlis, a high end restaurant in Seattle, switched to delivery back in March. You have to order in advance. According to an article in the Seattle Times, they have their waitstaff delivering orders. If you've ever dined there, the view is #1, the logistics are #2 and the food is #3, and the food is quite good.

A lot of high end restaurants in Seattle have done this. Copine, Artusi/Spinasse, Loulay, The Corson Building and, most recently, Cafe Campagne have been offering takeout with and without pre-orders. I get the impression that this is about "special" meals, the idea being that they are meant to be a treat. Amusingly Artusi/Spinasse offers its tajarin al ragu for $28, but only $10 if you cook it yourself. Even out in the sticks, a lot of high end restaurants were quick to offer takeout.

The low end restaurants have always offered takeout or drive through.

It's the restaurants in the middle that are bleeding. Granted, where I live almost as many had closed in early 2020 before COVID as have closed since.

Government charges you more if you use cash (legal tender) to pay tolls or fees. Pot calling the kettle black. Hypocrites all around.

This argument strikes me as a perfect example of economists simply missing the point. Why should delivery services commandeer 20 to 40 percent of a restaurant's receipts? That seems a clear case of economic rent: because they can. The cost should be put over onto the customers, who are insisting on delivery... or, and in many cities this is incredibly easy, the customers should walk or drive over to a nearby restaurant and get curbside pickup -- while, I recommend, tipping the restaurant 20 percent to help it stay in business under these extreme circumstances.

In this case, I would say both economists and selfish customers are both way off the mark. No one seems to care about the restaurants. I'm mainly thinking of local restaurants, not national fast-food chains... the restaurants that help define local quality of life. But economists would never consider that; it's not efficient.

I think missing here is the question of why should a restaurant charge the same for a take out meal as a sit down one? The restaurant's major expense is rent and the rent is driven by the sit down diners who:

1. Demand plenty of nice open space.
2. A nice part of town.

Sit down dinners take up most of the restaurant's foot print and cost them most of their rent. If they didn't have sit down customers, they could move to a cheaper section of town and just rent space for a kitchen.

Since the sit down people probably easily incur 40% of the costs, it actually makes a lot of sense that the app will take up to 40% and the restaurant still finds it profitable.

The app is just capturing a variance that the restaurant cannot easily capture. Since take out diners cost less, the restaurant would probably love to be able to offer a take-out menu with prices 40% less to get much more business but sit down customers would probably balk at such obvious dual pricing. The app, by keeping prices the same on the front, is just taking advantage of what the restaurant cannot.

Comments for this post are closed