Marina Krakovsky on the Middleman Economy
Mar 21 2016

Middleman%20Economy.jpg Why would anyone want to hire a middleman, like a wedding planner, especially if you have time to take care of the planning yourself? Marina Krakovsky, author of The Middleman Economy talks with EconTalk host Russ Roberts about middlemen in the modern economy. Despite predictions that the internet would destroy the need for middlemen, Krakovsky argues they're more valuable than ever though their roles have changed. Krakovsky looks at the different roles middlemen play today and how their value added can justify their existence.

Michael Munger on Middlemen
Mike Munger of Duke University talks with EconTalk host Russ Roberts about the often-vilified middleman--someone who buys cheap, sells dear and does nothing to improve the product. Munger explains the economic function of arbitrage using a classic article about how...
Robert Frank on Dinner Table Economics
How can you learn to think like an economist? One way is to think about what might be called dinner table economics--puzzles or patterns that arise in everyday life that would be good to understand. Robert Frank of Cornell University...
Explore audio transcript, further reading that will help you delve deeper into this week’s episode, and vigorous conversations in the form of our comments section below.


Mar 21 2016 at 3:55pm

It’s true that being a return customer, a wedding planner has more leverage on a florist than the bride/groom, but that’s just putting all their eggs in one basket because hopefully they are not going to be return customers with the wedding planner. There are also misaligned incentives. The wedding planner can get a kickback from its vendors, so she won’t necessarily use the option that’s best for her customers.

I’d like to hear the opinion on reservation from somebody in the restaurant business. Are no-shows such a big deal as this podcast suggested ? I’d assume that after 20 minutes you can cancel the reservation. Assuming that the average time that a customer is at the table is 80-120 minutes (high end place); that’s only 20% of capacity lost. And that’s assuming that the restaurant is full; otherwise there’s nothing lost. Places with a high percentage of reservations will have a problem because they have capacity loss due to the fact that reservations are inflexible. For example, imagine a restaurant is full and a patron leaves at 5:40pm. There’s a reservation for 6pm for just the size of table opened up, and it looks like nobody from a compatible table will leave before 6. So that table will be without guests for those 20 minutes. Then suppose that it’s a big party, so you set aside 150 minutes for the reservation. They might leave after 100 minutes, but you can’t seat anyone at that table if it is reserved from 8:30. Another 50 minutes of lost capacity.

Mar 22 2016 at 7:01am

Another aspect of the interesting insulator role is that sometimes anonymity between the parties is desired by one or both sides for completely legitimate reasons.

StubHub, for example, provides exactly that insulation between buyers and sellers. I recently had four tickets to a sold-out Broadway show, but we were stuck with a chance that one of my companions might have a last-minute conflict. Sure, if that happened, I could sell the ticket at the door: a handwritten sign or a quiet inquiry of buyer-eyed bystanders is a guaranteed way to scalp a ticket to a hot show. Even legally these days if you stay within certain price bounds.

But then you end up having to sit next to the buyer–someone who may feel buyer’s remorse and who knows for sure you were the seller. Awkward! Or, maybe the buyer himself might feel uncomfortable sitting next to someone to whom he has to feel grateful or beholden or with whom he has to make polite chitchat even if neither party wants that beyond the ordinary way you’d nod or say hi to the stranger you are sitting next to for several hours. Reselling a ticket through StubHub’s or Ticketmaster’s ticket resale options offers a kind of insulating anonymity that reduces social discomfort afterwards. The buyer doesn’t know for sure if the seller was someone seated on his other side or someone else altogether. An insulating middleman offers a protective culturally desirable shield between the buyer and the seller. Both buyer and seller may prefer a bit of anonymity because they have to rub shoulders after the fact. Literally!

Craig Brown
Mar 22 2016 at 10:40am


I thought this episode was particularly interesting in light of the Department of Labor’s proposed “fiduciary rule”, which is hostile to broker-dealers who earn commissions selling financial products for individuals’ retirement saving goals. These brokers tend to have incentives that are objectively misaligned with customers (e.g., they might recommend the mutual fund or annuity with the highest fees because they earn the largest commissions on such products). A financial planner who is paid a flat-fee or based on the percentage of assets under management is said to have interests that are more in line with those of consumers (although perhaps still biased toward over-saving and assets with high risk/return characteristics). Many certified financial planners market the value-add of unbiased advice in their promotional materials (books, radio programs, websites, whitepapers, etc).

It’s my opinion that we do not need a parental state to legislate how advisors or brokers are permitted to be compensated, for several reasons, the least of which being that the market already seems to be leading individuals toward index funds and fiduciary advisors in droves. However, I do not work in the financial services industry, and thus have two questions for you / Ms. Krakovsky / fellow listeners:

1) Despite holding political/economic beliefs which defend these middlemen and their freedom to employ commission-based business models of their choosing, as an individual investor I would certainly seek a fee-only fiduciary advisor to manage my personal finances. I am curious what distinct value-add you think can adequately explain the persistence of profitable middlemen who sell high-fee / poor-fit financial products on a commission basis to consumers? I imagine some will argue this is a terrific example of a market-failure of asymmetric information that can be fixed with sensible regulation. Would you disagree, and if so, on what grounds?

2) What might the negative unintended consequences of a mandatory fiduciary rule be? I’m generally thinking in terms of a thinner market which supports fewer products, less financial innovation, and provides advice only to affluent clients, because its possible that each of these might only be economic to offer (for better or for worse, depending on what you believe about the efficient market hypothesis and indexing) in the presence high fees and commissions. I also think it will be very difficult to programmatically determine what advice is legally in a consumer’s best interest because utility, costs, and risk are inherently subjective. What do you think?


Mort Dubois
Mar 22 2016 at 12:22pm

Interesting, but you’ve entirely omitted the shenanigans that middlemen get up to. I deal with both interior designers and general contractors – the former often takes a commission from their suppliers (that the end client doesn’t know about) and contractors are notorious for late payment or in some cases unilateral re-negotiation of terms of a deal, counting on the financial desperation of the counter party to prevent litigation . You also skipped over the tendency of politically powerful middlemen to legislate a permanent position in the market – car dealers, for instance. Without some acknowledgment of this kind of thing, the discussion seemed overly credulous as to the benefits of the monger.

And it may be covered in the book, but one of the primary functions of online middlemen, in particular Amazon, is to manage the logistics of packaging, shipment and delivery. That’s not necessarily a mediation function, but it’s essential in creating a very large market. Despite the ease with which the internet connects buyers and sellers, stuff still needs to be moved in physical space. Very large organizations have a big advantage in that respect.

Eugene Kernes
Mar 22 2016 at 3:03pm

Great discussion and examples of the cost of knowledge. The middleman usually already knows a lot of the vendors while the person who wants to consume the service or product will need to pay a huge cost in order to obtain the same set of information. It seems that the less frequent the purchase, the more value the middleman brings in as they have dealt with the infrequent purchase pretty frequently. On that note, the higher the frequency of purchase, the more knowledge obtained about what is good and bad about the product. As the middleman deals with the product more, they can understand which vendor/product is better for the particular customer than the actual customer.

The middleman provides knowledge via experience. The experience was gained only after huge costs. The middleman explaining the choice to the customer would come cheaper to the customer than the customer having to find all the knowledge required to make that choice. The middleman has a lower opportunity cost of the knowledge needed to make that decision.

Problems with middleman can exist like taking too much money than what would allow for competitive markets. The poor transactions occur everywhere, not just through middleman. Middleman that are well know can be trusted more than those unheard of because of reputation costs.

Mar 22 2016 at 6:12pm

Agree with Buzz – the restaurant business is way different than airlines and hotel. Airlines cannot re-sell that seat after departure or even before but too close to the departure time. Hotels must wait for late arrivals way beyond the time they can find anther customer. Besides, a table-hour is a lot less valuable Real Estate than a hotel room-night or airplane seat-flight. In addition, depending on the restaurant, raw materials and labor are significant so they don’t go to waste when you don’t show up. For Starbucks, a full shop attracts more people that can buy and leave or buy and stand. No one wants to sit at a table in an empty restaurant: “is there something wrong with this place?”

Günter Weinberg
Mar 22 2016 at 6:13pm

This is an off-topic comment so sorry about that but owing to the interest the host has shown on Bitcoin and the underlying technology making it work, that of the blockchain, and the interest followers of Econtalk also seem to have on the topic (myself included), I want to recommend a very clear and articulate explanation I recently came across on the subject. It is a talk Jason Griffey, a Berkman Center for Internet & Society fellow, gave on February 25, 2016, at the Metropolitan New York Library Council entitled “The What, How, and Why of Blockchain for Libraries”.


Mar 22 2016 at 10:28pm

A discussion in the best tradition of Econtalk: The subject–middlemen, a.k.a. intermediaries–are key to economic efficiency yet easily overlooked an often disparaged or reviled. The guest (and presumably her book) confounds conventional wisdom: far from eliminating middlemen, the “frictionless” economy expands their roles and opportunities and if anything increases their value. Her thesis also reminds us of the importance of liquidity, and the role of middlemen in providing it. The taxonomy of middlemen was also fascinating and original. Great guest and interview.

Mar 22 2016 at 11:50pm

One issue about middlemen that was left out of the discussion but is important is the agency problem. A broker/middleman can use his/her experience and relationships to get a better deal for a client–but could also instead solicit favors (kickbacks) from vendors in return for higher prices. The only person’s welfare more important to the middleman than their client’s is their own. Middlemen need to be policed.

Daniel Barkalow
Mar 23 2016 at 12:21pm

One thing the episode neglected to discuss was how customers find good middlemen, and how the quality of middlemen is enforced; why aren’t there just as many problems for the customer in finding a good middleman as there are in finding good suppliers? I think the answer is that customers can share amongst their own social networks reputation of middlemen more effectively than they can share reputation of suppliers. It’s a lot easier to find a married friend who wanted the same features in a wedding planner that you do than it is to find one who wanted the same features in a florist, and one who wanted the same features in a venue, and so forth.

A while back, my family got a bunch of exterior construction done by a general contractor who had managed a number of successful and pleasing projects in our neighborhood. We were able to get reviews of him from the people he’d worked for, and shared our experience with local homeowners we knew. This worked great, in contrast to the common perception of general contractors, even though we had some tasks that would be done by subcontractors who hadn’t been needed for other projects.

Middlemen don’t automatically make good guardians, but there’s the possibility to connect good guardians to suppliers through middlemen.

Also, to connect this episode to a different recent episode: I think one critical difference between the modern service economy and what people think of when they hear the word “servant” is the presence of middlemen in the insulator role.

Mar 24 2016 at 8:50am

I think that it is pretty obvious why restaurants can’t charge for no shows. When making a plane or hotel reservation, you know exactly what it is going to cost and your exposure (if you read the fine print long enough). Plus, they have your credit card at the time of the booking. When you go to dinner, you really don’t know how much you are going to spend – a feature that restaurants rely on to upsell you – and therefore you would not know the risk of not showing up. In addition, in order to charge you even a fixed fee for a no show, they would have to take your credit information in advance, and this is currently unacceptable.

One of the obvious middlemen mentioned but not discussed is Google itself. It’s search results are provided to maximize its revenue (like the other middlemen mentioned above who receive kickbacks from vendors, this is Google’s entire business model…), but if they also did not provide adequate concierge/reference services, they could not stay in business. The average American pays Google $10/mo for their middleman service – just because it is indirect doesn’t mean that it is free.

Mar 24 2016 at 4:24pm

There are a lot of middle-men not necessarily because the demand for esoteric information or because transaction costs are high, but because the demand for someone else to blame is so high.

#1 rule in selling for like 2000+ years is people ‘buy’ on emotion and ‘justify’ on fact.

Emotion is big money.

Middle-men act as lightning rods.

Regarding restaurants: no shows are missed sales and missed sales always hurt and are not easily substitute-able. Timing is everything.

Demand for seating is elastic with peak times and revenue following Pareto Principles with the highest demand at, say, Friday and Saturday at 7 – 8 PM. Staffing is expensive and has to be scheduled to accommodate demand.

The unit of demand, the person, comes in the size of ‘parties’ such as 2, 4, 6, 8, etc. and forecasting how these parties can and will be properly ordered and arranged, timed, has to be triaged and coordinated with the capacity of the kitchen to produce meals and with the amount of time diners utilize during their stay.

This process is always complex. There is a constant line balancing on the demand and supply side because the timings and units are erratic. Salmon cooks quicker than chicken. A party of 4 having 3 courses and wine turns over slower than a party of 4 having only 1 course.

Wait staff has to have their work-load balanced to insure proper service. 1 waiter to 4 parties of 2, does not equal the same work as 1 waiter to 1 party of 8. Etc. Cooking 8 chickens to be served at the same time is easier, but takes longer than cooking 1 chicken, 1 salmon, 1 medium rare fillet, and 1 pasta, which is quicker but more complex.

Open table acts as a lightning rod because people who use it aren’t going to complain to it. They take it at face value when they see no open reservations. Anything that diffuses hungry, angry, ignorant, emotional, or individual exceptionalism, in any industry is worth its weight in dollars. It can also act to compel some compliance.

Restaurants can’t enforce reservation compliance via credit cards because the credit cards company’s won’t support it.

Regarding wedding planners…

Why don’t we see PhD graduation ceremony planners, despite the commitment and sacrifice to that endeavor? Although that might be an untapped niche… If ever there was a diamond / water paradox of price/utility it is the emotional and financial capital spent on weddings.

David Longstreet
Mar 26 2016 at 5:59am


Since you are building up your own network of listeners and followers, it will be interesting to see how book publishers react for your next book.

It is a matter of time until book publishers/agents/publicist are calling trying to promote books.

In the end, Econtalk is a “middleman “for knowledge on economics.

Very interesting show.

Patrick R. Sullivan
Mar 26 2016 at 9:45am

Here’s a short video about a start-up in Seattle that wants to become the Uber of trucking.

Smart. Very smart. Having spent a couple of decades trying to find trucking companies you could rely on, I know this has potential.

Don Rudolph
Mar 26 2016 at 12:48pm

We recently took a trip to Thailand through ‘Affordable Asia’. I just realized they played the role of travel agent in a sense. They put together a package that would have taken hours to plan. The price seemed so reasonable I suspect some of the value we got as a customer was through their ability to negotiate lower prices. So the role of travel agent may have changed into the job of who ever put our package together.


Mar 27 2016 at 6:33am


Hear! Hear! Due to the circumstances of life, I had to pay for our wedding. I tried to explain to my soon to be bride (some 30 years ago) that we were just “going to church, having a party and going on vacation, why is this costing so much?”

I’m sure you know how it turned out. (It was a VERY nice wedding – at least she didn’t hire a planner.)

Mark Vaughan
Mar 28 2016 at 4:45pm

I felt the discussion about wedding planners earning a good living left out a key salient point; namely, that given a once, twice or three time-in-a-life event, the perceived cost of “failure” is high and pricing therefore somewhat elastic.

Secondly, there is also an observed (by me) competitiveness by new brides, to achieve the best possible imagined outcome. Given the wealth signalling role of a wedding embedded in the culture, one would also expect competition for the “best” wedding planner.

Steven Aves
Mar 30 2016 at 10:59am

Thank you Russ for another thought provoking podcast. This one in particular is near and dear to me; I’ve owned and operated my travel agency since 1984.

I wish I had time right now to share all my observations about being a 32 year travel intermediary, but this is my busiest time of year and boy am I busy this year! No folks, we travel agents are alive and kicking.

Middlemen and their value can not be fully appreciated without knowing the industry they are in and serve so as to avoid the “fallacy of composition.” Reading the comments above I can tell it runs amok.

Comments are closed.


EconTalk Extra, conversation starters for this podcast episode:

This week's guest:

This week's focus:

Additional ideas and people mentioned in this podcast episode:

A few more readings and background resources:

  • Information, by Joseph E. Stiglitz. Concise Encyclopedia of Economics. Discussion of the role of middlemen.
  • Ronald H. Coase. Biography. Concise Encyclopedia of Economics
  • Carl Menger. Biography. Concise Encyclopedia of Economics. Discussion of the role of middlemen.

A few more EconTalk podcast episodes:



Podcast Episode Highlights
0:33Intro. [Recording date: February 29, 2016.] Russ: My guest is Marina Krakovsky, journalist and author. Her latest book is The Middleman Economy: How Brokers, Agents, Dealers, and Everyday Matchmakers Create Value and Profit. It's a short book, but it's one that's rich in applied economics and ideas and relevance for today's world of commerce. And it's our topic for today. Marina, welcome to EconTalk. Guest: Thank you, Russ. Russ: Your introduction is titled--it's got a great title--"No One Likes a Middleman, but Most of Us are Middlemen." Let's look at both parts of that sentence. Why do you write, 'No one likes a middleman?' Guest: Well, this is a big theme in my book, is that people have long been suspicious of middlemen. People tend to think--and this is true across cultures, as far as I can see, across time--to think of middlemen as parasites. As people who aren't actually producing anything: They are not creating any value. They are just sort of living off the efforts of others. And of course a big part of my book is to challenge that view. Russ: Is it true? Guest: Well, it ¬can be true. It is true in some cases, or we think that it's true to some extent. But more often than not--you know, middlemen would not stay in business if they did not genuinely provide value. Russ: Yeah. I think that observation, and what your book does extremely well, is tell us what that value is. It's not obvious sometimes. But I think people have a suspicion that there must be some exploitation going on if somebody who is "not doing anything" is making money off the rest of us. I think that presumption--which I think comes unfortunately from a combination of there are such, occasionally, parasites, who are able to pull that off and with the help of, say, regulation that prevents certain things from happening; but also from just lack of understanding of what profits and the world of commerce is about. It does create a lot of, I think [?] and suspicion. So, why do you write: 'We are all middlemen'? Or most of us? Guest: Well, this is just the idea that--yeah, yeah. Middlemen are sort of hidden in plain site. I mean, anybody who is connecting with other people in a network is a middleman. Roughly speaking. [?] providing value in that process. And so certainly brokers and agents and dealers and salespeople--which, you know, certainly constitute a huge part of the economy, they fit the bill. But also people in non-obvious middleman jobs, like, wedding planners. Or people you know who are middle managers in a company, who are transmitting vestiges and coordinating the efforts of multiple people. These are also middlemen. I mean, if we really stop to think about what we do and the role that we play in our own social network, we are all middlemen. And the question really becomes: What value can we create? What more can we do to facilitate productive activity within our own networks? Russ: I guess in some sense I am a middleman. Even though I don't charge my listeners. It's through, Liberty Fund, the sponsor of this podcast. And unfortunately I don't pay my guests other than through the excitement of reaching a broader audience. But you could create your own podcast and reach out to people who are interested in your ideas, or try to interest people in your book. But you are using a middleman. Guest: Yeah. Exactly. And that's actually why middlemen have become more prevalent with the rise of the Internet. The Internet has reduced distribution cost, has enabled many people to take economies[?], traditional gate-keeper roles: instead of a radio station or TV network or magazine deciding who to have on as a guest or whom to interview, now it's anybody who is running up a podcast or a blog, or the like--is playing that same role. So there are many, many more people playing those middleman roles. Russ: And as a result, the middlemen who are involved in that process--of publishing a book; used to be there were a handful of places you'd try to get your author on. And now, it's a much broader list. It's a much more challenging, and I suppose, game than it used to be. Guest: Yes. And it's always changing.
5:03Russ: So, it's interesting that the Internet has increased the potential for intermediaries. You quote Bill Gates, that people would originally--I don't know which part of this is Bill Gates. His part is the phrase, 'friction-free capitalism.' But some people, maybe it was Gates, but certainly others thought that, once the Internet comes along, we can all look for stuff; we are going to have friction-free capitalism now; search costs are almost zero; we're not going to need middlemen any more. No intermediary is--just go straight to the source. And that certainly has not happened. Why not? Guest: That's right. I mean, the essence of it is that transactions costs have fallen for everybody. But they have fallen more in most cases for professional middlemen than they have for the rest of us. And so, if professional middlemen can do their work more efficiently than we can, then it still makes sense for us to turn to the professionals rather than doing things ourselves. Russ: So, let's take an example. One of the more mundane but fascinating examples that you mentioned a minute ago was the wedding planner. What role does a wedding planner--how do you make a living as a wedding planner? I mean, there are people who do this for their friends, or used to do it for their friends--they'd go out and try to help the bride or the groom get stuff done that was needed for the day of the wedding. But now there's a very specialized task called a wedding planner that is persisting in the modern world. How did it come about and why does it persist? Guest: Yeah. Well, wedding planners have been around for a long time, but they were seen as a luxury and something that only people who spend a fortune on their weddings can afford. And now more and more people are hiring wedding planners. And most of us, we think about a wedding planner, we think about a coordinator, someone who just handles all the logistics, all the hassles of planning a wedding. But a wedding planner actually does a lot more in this middleman sense. So, I see a wedding planner as fulfilling a couple of main roles. One of them is the concierge--the person who is handling all the pesky little details and making the process of planning a wedding as simple and carefree as possible for a bride. And also helping the bride make all these high-stakes decisions without getting too anxious about it. So that's the concierge role; and I think a lot of people can relate to that. But I think, more interestingly, a wedding planner is also an enforcer. And what I mean by an enforcer is the person who has the clout to keep both sides honest--in this case, it's primarily keeping the wedding vendors honest. You can get the same list of wedding vendors that a wedding planner might use. But those same professionals are probably not going to do as good a job for you as they would for the professional wedding planner. And there are various reasons for that. One of them is just that the wedding planner can speak the language of both sides, because of her--typically it's a she--experience planning weddings. She knows how to talk to florists and bakers, and all the other--photographers, and so on. And understand what brides want. And so it's sort of a translator, a cultural broker between the two sides. But more importantly, I think, the wedding planner just has the long-run reputation and long-run relationship with all these vendors that she can leverage on behalf of the bride. So, a bride is not going to be a repeat customer, and therefore won't be able to get the same level of quality and service working directly. Russ: So, I thought this was a fantastic example--again, it's pretty mundane--but it really highlights the economics, I thought really beautifully. So, in a way, you list 6 different types of middlemen and they overlap, as you point out, also--and we'll probably get to all six. But the enforcer is one of them--somebody enforces quality. Concierge is a second. But you think of a concierge in the modern world as being somewhat obsolete--because, you know, 'Hey, I can use Google. I can look up florists; I can see how they are rated.' And I thought the point you made--first of all, sometimes that list is very long. And sometimes there are things about florists that you don't know, that doesn't come up in the reviews, that are very idiosyncratic relative to what's important to you, this venue, the way you want them arranged--whatever it is that the wedding planner can do and then pick from that list in a way that you can't. But I think the deeper point, the one which I love, is the one that you made, is that even if you had the list of who are the really good ones, say, even there might be some ones who might have good reviews who aren't good, or who are the ones who are right for you--how they will perform is not the same. And I think people forget that effort is really important. And that any one time, there's a potential for slacking. And not doing the best job. And that the long-term perspective of that wedding planner has a huge impact in terms of the incentives facing those vendors. Guest: Exactly. You put that very well. It's really that distinction between adverse selection and moral hazard--sussing out the intrinsic quality of somebody, their potential to do a good job versus their incentive to actually do a good job on any given transaction. And the long-run player has an advantage over the one-time buyer in eliciting the highest quality service at any one time. Russ: And there's a real art to the concierge part. As you point out, a lot of people now, they don't want to see a person. They are happy to sit in their hotel room--the literal concierge--google around for restaurants or sightseeing or things they want to do. So, how does a human concierge compete effectively--I give one example: sometimes you'll know that a rating was earned a long time ago; they've lost their qualities and what it used to be; but a really first-rate concierge can still thrive in today's world. How is that possible? Guest: Yeah. I mean it happens because there are these individual differences. Some people really do prefer to do things by themselves, and they don't mind spending hours and hours figuring things out for themselves, like a big trip that you are planning. Maybe that's the classic example. If you are planning a simple trip, yeah, it's very easy to do this yourself and it doesn't take much time. You jump on your favorite website, then you make your bookings. And you can do that very quickly and easily without much angst. But if you are planning a complex vacation to a destination you've never been to, there are a gazillion choices; and if you are the kind of person who doesn't really enjoy doing hours and hours of internet research, and if you are the kind who gets anxious about making the right decision or the wrong decision, then you are the perfect candidate to work with a concierge--in the sense that I am using 'concierge': the travel agent who specializes in your destination, who has planned these kinds of trips for many, many past clients, who is very good at matching your deepest needs and wants with the kinds of options that are available in that destination. They can save you a lot of time and self-doubt about making the wrong decision by planning out an itinerary that is pretty well customized to what you want. Russ: Yeah. I took a trip to London last year. I'd never been to London. And there's 80 hotels in the area that I'm pretty sure I know where I want to be in terms of geography, which part of London. But there's 80 hotels that have--pretty good ratings. And I'm not convinced they are all going to be fine. Sometimes they are all fine. Right? Guest: Yes. Russ: But all of them have negative reviews; and you are always wondering: Was that just a crazy person who didn't like the orange juice? It's scary. Guest: That's exactly right. Yeah. Actually the travel agent that I interviewed for this book said, 'You never know who the person is, if it's somebody in their underwear sitting in their basement writing their review.' So, yeah; she can suss out those kind of things. But even if you know that all those reviews were good, there's still that matching process and there's still the question of: How do you factor in, how do you weigh all these various factors--the location and the price and--oh, gosh, there are just so many aspects to figure out--so that one decision can easily bog you down. And so this is where it's really helpful to have somebody just kind of make that decision for you. Russ: Yeah; I was impressed there are still travel agents. Because I don't use one. It's an interesting thing. I don't travel to London often. I can imagine going again in the next 5 years. But I'll probably go back to the same hotel; and that may not be the best thing. But it will be fine--since it was okay, I'm probably going to go back to the same one. But I think what's interesting is how hard it is when you are doing something infrequently--ideally, getting married is something you only do once; you might do it two or three times, but you probably won't do it--even if you do it more than once you probably won't end up doing it in the same city. So this opportunity to rely on somebody else's expertise is really pretty fantastic. Guest: Yeah. I agree. I agree. And it's underappreciated. After writing this book I've started outsourcing more of those kinds of tasks that--I mean, the thing about the Internet is, it's very easy to feel, 'Yes, I can do it myself, so why should I ask somebody else to do it?' But it's just so easy to dive in and drown, as somebody put it. Russ: And it's worth pointing out that even when you use the Internet, often you are using a middleman. So, I happen to use Kayak a lot. Kayak is a middleman, right? I don't go to each hotel's website; or TripAdvisor--I don't go to each airline or each hotel's website. I use an Internet intermediary. Guest: That is so true. Yeah. The Internet is full of middlemen like that. And the question is: Well, which middleman do I trust? And how do I compare them? And it's just very easy to get sucked in.
16:03Russ: So, let's talk about an example of an intermediary on the Internet, a middleman on the Internet, which is Open Table. So, Open Table--again, as an economist, really interesting example. Talk about what their business model, how it started, and the challenges they faced and how they overcame them through the decisions they made. Guest: Yeah. Open Table is an interesting example to dive into because as diners, we sort of take it for granted. It seems so simple. But there's actually a lot going on behind the scenes to make it all work. Because it's a two-sided market. And with two-sided markets you always have this challenge of getting both sides on board. Because Open Table would not be valuable to diners if plenty of restaurants were not on it. And so the first challenge--so even if you give away the service for free to the diners, there's still the challenge of getting restaurants to sign up and to pay for being on it. And restaurants pay a monthly fee; and they also pay a booking fee that is about a dollar typically for every seat, for every person in the party making the reservation. And it was actually really hard to sell this idea to restaurants, even though restaurants stood to benefit a lot--I mean, they got marketing as a result; and the big selling point is to be able to fill in seats at the last minute. Russ: Just to make it clear for those who don't know: it's a reservation service. To make reservations at a restaurant. Which seems like kind of a trivial thing. Why would I need to get on the internet? Why can't I just call the restaurant, make a reservation? It's bizarre on the surface that such a thing could make a profit. Guest: Yeah. Yeah. So, to step back: Yes, this actually solved a big problem for diners, which is that you are trying to make a reservation, and typically you have to call--without a service like OpenTable, you have to call restaurants one by one to see if they have a table available for your-size party at your time. And it can be a pain. If you've ever been through this process. Sometimes you have to leave a message and hope that somebody will call you back; and often there is no availability. But if you go to a site like OpenTable, which you can also get an app for, then with just a few taps, you get to see right in front of you who has openings for when you want in the area that you want. So, it greatly simplifies that search process. So that's it from the customer's point of view. Russ: How about the restaurant? Why would they bother? Guest: Yeah. Okay. So, for them, there are a few benefits. But the biggest, and it basically boils down to filling those tables. And also not having to answer the phone as often; not dealing with that appointment book, even though most restaurants will even take reservations over the phone, even if they are using OpenTable. And so, the promise to the restaurant is that they will have fewer openings, because diners will be able to see, 'Oh, look, here's an opening at 7 o'clock. I can't get a table anywhere else, but I can get one at this restaurant.' And boom, you've got your booking and the restaurant has a table filled. So that's great for both sides. Russ: The challenge a restaurant faces--which was the interesting part, I didn't think enough about from the value proposition that OpenTable is providing: You've mentioned some of them. But one of them is the fact that people--it's kind of obvious--not everybody keeps a reservation. And so, a last-minute cancelation or a no-show--a simple no-show, they make their decision and don't show up--can be disastrous for a restaurant. It's very expensive to leave that table unfilled, to have held it and then not fill it. And so, the obvious way to do that, to avoid that, is to charge people to make reservations. But that has a problem, too. So, talk about the problems that that has, and how Open Table manages to reduce no-shows and cancelations. Guest: Yeah. Cancelations are not as big a problem for restaurants as no-shows, because if somebody cancels before they show up then the restaurant can still take a walk-in, potentially at least. Whereas if you made a reservation and you just don't show up, then you're like a dog in the major [?]--you are holding on to that table and the restaurant can't give it away. But, at the same time, the restaurant hasn't taken your money. It's kind of a funny thing, because it's part of the hospitality industry. It's just different parts of the hotel and hospitality industry work different. So, hotels take a non-refundable deposit--they take your credit card, so if you don't show up, your credit card still can get charged. But for whatever reason, it's traditionally been a no-no in the restaurant industry to charge diners. And if you were the one restaurant that sticks its neck out and tries this thing, well, that's going to hurt you. So there has to be some kind of change, a collective change. And that's starting to happen with higher-end restaurants in big cities sometimes. But when Open Table was starting out, this was basically unheard of. And so restaurants could not charge for a reservation. They couldn't do it themselves. They couldn't do it through Open Table, in fact I don't know if Open Table even considered it. So they are stuck with this problem of No Shows. And No Shows can be a problem online or they can be a problem over the phone. It's a perennial bane of restaurants' existence. But their fear was, when Open Table came to pitch their service to the restaurants, the restaurateurs feared that no shows would actually be more common if people could make their bookings through Open Table. And their concern was that people just would not take these online reservations seriously. That OpenTable was making it too easy to create a new account. You know, you could become John Doe at, or John Doe at ThreeFiveSeven,, or just to have multiple accounts and make reservations and then no show within impunity. And so, OpenTable, to convince restaurants that this wouldn't happen, really had to create policies to minimize the chances that this would happen. And they did a bunch of things about that. And it's very tricky to do. Because they didn't want to make you sign up with an official email address. They did want to make it easy for diners. So there's always this need to keep both sides happy and they managed to do that. And they did it in a number of ways. One of them was this loyalty program, which gave people an incentive to kind of just stick with one email address, because they would get rewards. They would get these points for every reservation that they made, and actually completed. And they would get a cash payout after a certain number of these. So they wanted to stick with the same email address. And also, the restaurant, Open Table, would remind people of their reservation, a day or two beforehand. Which is such a simple move. But it's really helpful because people just don't think about the restaurant--I think about this--but most people probably don't think about this: people who routinely no-show are not thinking about the problems that their no show is causing the restaurant. So, it's not necessarily just that they are being selfish or malicious. They are just not thinking of it. And so that little reminder email often is all it takes to get people to show up. And then, if they don't show up, the OpenTable also keeps track of that. So, after a certain number of--I just wanted to say that after a certain number of no-shows they get kicked off the site. So, you know, there are consequences. And these are kinds of consequences that Open Table as a middleman is in a better position to impose than any restaurant could on its own. Because they are aggregating data from across restaurants. Right? So they know that I have not showed up at this restaurant and that restaurant and the other restaurant; whereas only one restaurant would only know about only that one no show. And would also have a harder time from just barring a customer from making a reservation in the future. So it's just one of the many ways that a middleman has an advantage over the buyer or the seller interacting directly. Russ: And I think--don't they send out--it's kind of thoughtfully worded--if you miss the reservation, so it's not like: 'How could you do this, you awful person!' but 'Did something go wrong? Do you remember what that is?' I can't remember. Guest: Yeah. I mean, the interesting thing about that is that Open Table is not on site. There's not like there is a company representative at each restaurant checking people in. So, they have to go by what the restaurant is telling them. But at the same time, they know that the restaurant might not always be honest with them. Because the restaurant has an incentive to report a no-show. Just a small incentive to cheat that way. Russ: It's their dollar. Guest: Yes. Russ: It's a lot of money. Guest: [?] You know. And restaurants quickly realize that it's just not worth it; and Open Table will be on to them. But I mean what's really clever about this is that Open Table is able to police all of this from a distance. They are really asking the diner to then say, 'No, no, no; I was there. What are you talking about?' And that's how they can resolve these disputes. And of course if the restaurants realize that there's this loop, that the diner will have a say in the matter, then they quickly learn not to play those games.
26:37Russ: So, I just want to speculate for a minute, because you talked about the cultural phenomenon and hospitality industry. You stay in a hotel, they usually give you a rule that if you don't cancel within a certain time frame, which is usually 24 or 48 hours, they are going to charge your credit card. Of course, you know that in advance. They tell you that. The first time, you may not know it. But you eventually learn it. Whereas other types of no-showing, like restaurants, typically have no costs. And one argument would be: Well, you don't know how much you are going to spend at the restaurant. Whereas the room is fixed; you know what kind of room it is. But a restaurant could charge a flat fee--$10, $20--and so just to think about it for a minute--it's an interesting dinner table economics question for those of you out there who--you might want to pause this and think about it on your own for a minute; I'm just thinking about it here; I'm going to give my answer and then I'm going to let Marina respond. But it seems to me that, of course, we do pay for the no shows. They just get spread across everybody. It's part of the cost of doing business. It's somewhat akin--not exactly, but it's somewhat akin to somebody walking out on a bill or stealing something out of the hotel, in the case of a hotel like a towel or something else. It's part of the cost of doing business. Every restaurant knows that some reservations won't be shown--people won't show up. And so they can't maximize their capacity on any one night. And since they have to make a living and cover their fixed costs, a little bit more to make it worthwhile, presumably the prices on the menu reflect that. So, you are paying--it's just in a very hidden way--that gives you the flexibility of canceling without a direct cost. It's built into the cost of every meal in most restaurants that take reservations. So then the question would be: Why doesn't a restaurant come along and charge lower prices on their menu; but reservation fees, so that people who are honest and keep their word would be able to enjoy cheaper meals? And I guess the answer there is that most of us want the flexibility of choosing where we eat on any one night. We might want to want to make a reservation a few hours in advance or a day in advance or a week in advance. But we like to be able to change our mind. Or something comes up--much more often than a trip to a faraway city gets canceled, the hotel reservation. So I think that might be part of the reason why that cultural norm is there. What do you think? Guest: Yeah, that's an interesting analysis. And you just reminded me of something else that, it's very clever, that Open Table is doing. It's so simple. When you said that people want the flexibility of making a spur-of-the-moment decision about where to dine: People in the early days would make multiple, would make multiple reservations for the same exact time--which is impossible to fulfill. And so one of the rules that OpenTable instituted early on is that you cannot make two reservations within two hours of one another--such as simple technological fix to a problem. Um, so that was interesting. Yeah. I don't really know why the cultural norm exists. And of course there are exceptions to that. If you want to make a reservation for a New Year's Eve Party, a prefix, you know what the price is, you have to put your credit card down. Or some popular restaurants in New York City, or there are large parties, and so on. The restaurant is always trying to balance the cost of, you know, the possibility of a no-show, with the cost of, you know, disgruntled customers. I think that the norm is beginning to change very slowly. But I really can't say why it's been the way it's been the way it's been. Sometimes these things just sort of evolve for no good reason. Russ: Yeah. We've talked on this program before for the markup for food, contains the fact that you sit at the table for a while--and this was--Earl Thompson, who was at UCLA (University of California, Los Angeles) and made the point that the longer it takes to eat something, the more expensive is the restaurant, because it takes a longer time for the table to turn over. And that means the number of customers per night is going to be lower. That means you have to recover your fixed costs from a smaller group of people. And therefore the prices have to be higher. It's just a--I think it's a fantastic insight that captures part of what's going on. Guest: That's right. I just want to add to that that when you do takeout at that same restaurant and you are not costing them a table at all, you still pay the same price. I don't know any restaurant that charges less for takeout than for dining it. Russ: Well, actually some do. Guest: Okay. Russ: In fact, Emily Skarbek at EconLog, a sister site of EconTalk under the Library of Economics and Liberty, is recently speculating about that. Of course, some take-out prices are cheaper. But of course then you have to monitor people who do the takeout thing and then go sit down. Guest: Right. Russ: Which is one of the challenges of charging different prices. Or, I was going to say--I lost my train of thought--but it's back, is that: Given that it takes longer, one solution would be to charge, have a meter, at each table, how long you are there. If you want to dawdle over your coffee, you can. Or your dessert, or whatever it is. Of course, people--that somehow takes some of the fun out the dining. The clock's ticking on you, I guess. Guest: It bursts that bubble of romance around the dining experience where you feel like, you, yeah, you are a guest at somebody's house and you have to pay at the end of the meal. It's not always in your face. Russ: And there are guests who probably should have a meter in your own house, even. But we'll leave that to the side. But a serious thought is that Starbucks lets you buy, lets you in the sense that there is no explicit rule against it--you can sit and buy a cup of coffee and use it as your home office for hours-- Guest: yes-- Russ: and they never say, 'Can I get you anything else?' And meanwhile, people who come in, would like to sit down for 15 minutes can't find a table. It's a very interesting decision they've made to not charge, to not only not meter how long you are at the table but not encourage you to buy more stuff to earn the right to sit at that table. It's basically rent-free, all day, with the purchase of anything. And even, I suppose if you just sat down, I think they would probably leave you alone, too. Kind of like a public library, with coffee on the side. Guest: Yeah.
33:05Russ: I want to turn to an issue that you discuss in the book that's really fascinating, which is: Some of these jobs--and the wedding planner is one of them--and to some extent OpenTable is another, but the wedding planner is a particularly good example--the skillset of the woman that you highlight in the book and feature in the book, in a way there's nothing explicitly obvious about what she's "good at." She makes a very good living, evidently. She makes a relatively large amount of money. And does this, not just--as you point out--not just for high-end fancy weddings, but for more average-cost weddings. People willing to pay for--not trivial amount--to get her assistance. How does that persist? Why doesn't competition--there are no barriers to entry in her case--there usually aren't in these kind of situations. What's keeping her fees as high as they are? Guest: Yeah. This is a question I worry about [?] in the book. And I think it comes down to being able to make increasingly good matches the longer you are in business. And there's an interesting study I talk about that deals with placement agencies and how they are able to make better and better matches. And these economists actually were able to analyze the numbers, and could find that--well, I don't remember the details but something to the effect that the more that a given client--well, actually not a given client, but a worker, a particular agency, the higher the markup that that agency can charge for that worker's service--for that worker's time. And furthermore, the more of that markup that agency can keep--so the agency benefits more from the longevity of that relationship than does the client, than does the worker. So, what's going on with the wedding planners is somewhat similar I thin, which is that the longer that they do business with the various wedding vendors in their town, the more they know, you know, which florist can make a beautiful bouquet of daffodils or orchids or whatever it is. And which women, provider, you know, happens to have a silver tablecloth. I'm not very good at coming up with these examples. Russ: I can't help but mention that when somebody tells me they are getting married, I always say, 'Remember that no one remembers the color of the napkins.' And the amount that people agonize over these decisions always depresses me. So I'm just going to say that. But carry on. Guest: But yes, yes. But these are all things that are all very, very important to the brides. And having been a bride, I understand that. It's the most important day of your-- Russ: Some grooms also care about the color of your napkins. I just wasn't one of them. Guest: --life-- Russ: I just want to say that. Guest: Okay. So, going back to the wedding planners, they know who to go to. They can do this very efficiently. And they can do it with increasing efficiency and increasing equality over time. And so it's hard for a new wedding planner to compete with that. Having the reputation of course for doing all of that also makes it easier for the established, just like in any business, for the established wedding planners to, you know, to get the new business.
36:50Russ: I never thought about that with the employment agencies. It stuns me when I read the markups that employment agencies are able to earn. And some of that is due to the fact that they absorb some of the costs of a lawsuit; they absorb some of the costs of unemployment. They, unfortunately are not--the regulatory stuff around laborers in working has gotten more expensive, which means that it's more and more profitable for somebody to take on all those costs in helping--not just in suggesting workers, but in being the employer for, in setting up people as temporary workers; but they work for the agency, not for the client. But the point that it's always difficult to ascertain quality, and that I can't just tell an employer that I'm a fabulous worker, but that somebody that vouches for me and whose reputation is on the line--the ability, to have the ability to create a brand name around something that doesn't normally have a brand name, or a brand name is hard to establish. Guest: Yeah. That's exactly right. Now you are talking about what I call the Certifier role: the ability to credibly certify because you are staking your long-term reputation on your recommendation. That is one of the unappreciated or underappreciated values that a middleman brings to these transactions. So, a placement agency is an example of that. And I was surprised to see it--among other surprises in the book--that that's happening on Craig's List and e-Bay, which, I'm not a big consumer or provider of stuff on those sites. The idea that there are people reselling stuff on Craig's List, use stuff--now, one of the reasons, I guess is that it's hard to brand name for used stuff. It had a brand name once, when it was new. You don't know what it's gone through; so you don't know quite what the quality is. You can take a picture. Which is nice. It helps. For some goods that's enough. But for other goods, it's not. Talk about people who are making a reasonable living, or at least have been, on Craig's List, reselling other people's stuff, that's available on Craig's List. Which was amazing to me. Guest: I know. It's so counterintuitive. Like, can anything be more direct than Craig's List? Well, I guess you can think of other things, because Craig's List itself is a kind of middle man. But why do you need additional middlemen? Russ: Well [?] the middle man? Guest: Exactly. And it's not just on Craig's List. It's on E-bay, it's on LinkedIn; it's on YouTube. It actually happens all over the place. Middlemen emerge to serve these various needs. And I think the biggest need is that the middlemen can fill on Craig's List, and this is a person I profile, is to provide liquidity to both sides: you know, the person who has an appliance that they are eager to get rid of because they are moving or because they just need quick cash; they are in desperate straits; and they just want to get rid of it as quickly as possible. So, they are willing to take very little money for it. And that's what the middleman on Craig's List can do. He can swoop in as soon as he sees it, and he's the first one there; and he takes that appliance off that person's hands. And at the same time, he's able to provide that immediacy to the would-be buyer, because he's not leaving that price artificially low. You know, he gathers up the inventory when prices are low--which tends to be cyclical. And then he puts it out there at a price where it won't get scooped up necessarily the same day; but because of his experience he knows that it will get bought at that price within a reasonable amount of time, and he's willing to sit patiently and wait for that to happen. And that provides a service to the buyers because there's a good chance that an appliance will be available on Craig's List when the buyer needs it. Russ: And that person is actually taking the appliance into his house. It's not like one of these paper transactions where he is listing it--so that's what is interesting about it. He's not just creating his own list of what's for sale on Craig's List. He's buying it; and then reselling it; but he's actually receiving it in the meantime. Right? Guest: Yes. Yes. He does buy it outright. But it's very cheap. So, you know, it's not like a huge capital investment. He holds onto these appliances in his garage; he can 15 or 20 washers or dryers there. And then he flips them. He turns around and sells them within days. Russ: And he makes--is this a fulltime living? I think it was, right? Guest: Yeah, it's a full time job and it's a solid middle class income. He reports making $1200-$1500 a week doing this, profit. Russ: It's a lot of money.
42:04Russ: The other thing I was shocked about is the Powerseller role on e-Bay. I think you said 4% of the sellers on e-Bay are half of the merchandise. Is that--do I have that number right? Guest: Yeah, that's exactly right. More than half of all sales on e-Bay flows through what are called Powersellers. Which are essentially middlemen, because to get that Powerseller designation, you need a certain sales volume, and a certain minimum feedback score which is rather high. So in order to achieve and sustain that level, you need to be basically a middleman, buying and selling other people's stuff. So, most trade on e-Bay does not happen the way that it was originally billed. It's not--dispenser collectors trading with one another. A lot of it is professional middlemen buying, oh, somebody else's user designer fashion is the example that I use; and then selling it to people all over the world who are interested in the same stuff. Russ: It's like running a perpetual garage sale where you are always buying stuff from other garage sales and you've got a reputation for throwing out the junk and keeping the better stuff, so it persists. It's amazing. Guest: Yeah. Russ: Let's talk about a hated middleman, which is a used car dealer. A lot of people resent, when they go to sell their car, they notice that if they sell it themselves they can get a much, much better price than the dealer will give them, and they tend to say, 'Well, it figures; dealers are awful.' But explain why there's that gap to start with, and then how a middleman named CarLotz , a company called CarLotz , is trying to change that process. Guest: Right. Right. There are really three sets of prices. There's a dealer trade-in price; there's a sell it yourself price; and then there's the price that a dealer could command. So if you are not a dealer then you are really choosing between trading in your used car--I'm talking from the point of view of someone who is trying to sell a car now--you can trade it in to a dealer and get much, much less than if you sold it yourself. Those are your two main options. And neither is very attractive because they both have huge downsides. It's a huge hassle to try to sell a car yourself. But if you sell it to a dealer then you are going to lose a lot of money. So, what CarLotz is doing is they are trying to step into the middle, in this sense, the middle of the market, to give you kind of the best of both worlds--the benefit of selling it yourself, meaning that you get to keep a lot of the value of the car, but also the benefit of selling to a dealer, which is that you don't have to deal with all the hassles. And their model is to make the process as efficient and as transparent as possible. Which is what the traditional used car sales process has lacked and why so many people hate used car dealers. And to address your question of why there is this gap: So, part of it is the lemons problem--the problem that the owner of a car knows more about the quality of the car than the buyer does, and so you kind of have to assume that the quality isn't very good or you have to pay a mechanic to inspect it; and there's a cost to that. But a bigger problem according to this guy who started CarLotz, is that when you trade in your car, you are just at the very beginning of a very long inefficient supply chain, and so your car has to go from a dealer to a wholesaler to an auction house to another dealer, and there's transportation and there's a time lag, and so a car is depreciating. And so there's a lot of leakage throughout the process. Russ: So, what is CarLotz doing differently? Guest: So, CarLotz buys--well, it actually doesn't buy your car, first of all. It takes it on consignment; and then they, instead of trying to--so, there are a few things. They take it on consignment rather than buying it. They charge a service fee for selling your car: instead of profiting from buying low and selling high, they charge a service fee to both the seller and the buyer. And that fee is fixed regardless of what they sell the car for. And they also don't pass that car on to a long chain of other intermediaries: they sell it directly on their own lots. Russ: So, I guess the argument there, what they are trying to--again there's sort of the business proposition is that I don't have to be home all the time to meet the prospective buyer of my car. They get good at vetting buyers' reliability. So, one person shows up at my house and I have to take their check or I'm not sure--it takes a while, they've got to get a bank check. So they can get better at, say, credit and other things. And they also, they [?] have lots of cars. So more people are going to come there to economize on the search costs. So, like you say, it's sort of the best of both worlds. Guest: Yeah. It really even comes down to something as simple as posting pictures of your car. Now, most of us, when we sell our car we can maybe take 5, 10 pictures of our car--that just seems like a big deal. It's not something we do all the time. Whereas, they'll take--typically they'll take 85 pictures; and they know exactly what angles they need, what customers will be looking for. And they do it in a matter of minutes; it's like an assembly line for them. So they can do it very efficiently. And then, whereas we might post it on Craig's List or maybe, they post it on 10 sites. So, they are just better at doing everything that you could on your own; they just, this is their prose[?]--this is what they do. Russ: And right now they're in one city? Is that correct? Or they are expanding? Guest: No; they're--yeah, they've expanded. Even when I was speaking with them a year or more ago they were already in several cities in Virginia and they were expanding out to a few other states. Russ: So, CarMax does that as well. But I think the part that is interesting about them--if you go to CarMax, you get an estimate for what they'll pay for your car. Which is very nice because it's real--it's an actual amount. It's not complicated like the deal, whether you are doing a trade-in, you're not sure whether the value of the car is actually what you are paying for the new car. But here I guess the other part, the novelty of it is just the fixed amount--and it's the fixed amount, the consignment aspect of it is fixed amount per car. And as you point out, that encourages people who want to sell higher-value cars in general. Because obviously if it's too high amount extra then you still might want to sell it yourself. Guest: Yeah. Yeah. They charge a fixed fee regardless of the final sale price of the car. And what's also interesting about that is, because they charge a fixed fee, the buyers also take that as a signal that these dealers, these salespeople, aren't pushing a particular car on me. They are really in the service business of trying to help me find the best car for me because they have no incentive to push any particular car.
50:29Russ: So, I don't want to miss your taxonomy here. Your six categories are: the bridge, the certifier, the enforcer, the risk-bearer, the concierge, and the insulator. I think we've talked about three of them; and really the bridge I think we've talked about implicitly, so that's four. But talk about the risk-bearer and the insulator. Guest: Okay. Well, so the risk bearer is very much related to being a certifier or enforcer, but it's all about really trying to pool the risk and discern external risks from what I call internal risks, which economists would know as counterparty risk. So, if you are a certifier or an enforcer, you have to know that the parties you are dealing with are trusted. That's what I would call the internal risk. And you want to avoid internal risk. You want to avoid dealing with people who are known for sub-par quality or who you can't rely on. But you do want to embrace what I call external risk--the exogenous risk that we really cannot predict: which startup company is going to take off and become the next Google or Facebook, venture capitalist example. That's the kind of risk that they are in a position to embrace, because they are--the building of [?], right--they are pooling, they are aggregating risk from across many sources. So they are in a better position to handle, to bear that kind of risk than someone who is just doing individual sales. Russ: Talk about the insulator. Which I love. The insulator is the greatest. Guest: Thank you. I think the insulator is so cool. And it's one that even economists don't really talk about all that much. The insulator is--it's very different from the others, because the others are more about facilitating connections that aren't there. Whereas an insulator is someone who keeps people apart in strategically important situations. So when we have an existing relationship, sometimes it's actually more difficult so say certain things to the other person. For example, there's often this tradeoff between, like, ability and assertiveness. You might be better off in the short run to assert yourself, but you might seem less likeable to your trading partner. And so this is for someone like a sports agent or a lawyer can step in very shrewdly to intercede on your behalf with your team or your client. Publicists play this role with the media on behalf of their clients because tooting your own horn, and other things that are sort of unseemly to do. And so the insulator can be an important role, but it's often invisible. We don't appreciate it. Russ: What I find is fascinating is that obviously, middlemen put people together. So the idea that you keep them apart--so you do put them together, but you do it in a way that's at arm's length thing. And I think people underestimate--again, they look at certain middlemen and say, 'Oh, they are a waste of money because all they do is take a fee.' And of course they also have to represent you at an arm's length, which means they might way things--just like it's unseemly for you to say, 'Pay me more money,' socially, it's also unseemly for the agent to not ask for enough; or to ask for too much-- Guest: Exactly. Russ: So there's a real risk, but it's worth it in many of these cases, obviously. Guest: Yeah. It's such a win-win, because what is bad for you is actually great for them, so if you are a sports agent and you talk tough on behalf of your client, well, you enhance your reputation as this great advocate, and so athletes flock to you. So it's almost exactly the opposite of what you wouldn't do for yourself because it would be bad for your image--well, it would be wonderful for the middleman's image to do those same things on behalf of another person. Russ: I was actually making a different point. I was actually saying that the agent has to be really good at figuring out what the client would want in terms of both being aggressive and easy-going, because I'm not there. I'm not in the room. Sometimes. Guest: Oh, I see. Russ: So the agent is talking to the press about the client, and the client can sometimes be grimacing, like, 'I wouldn't have said it that way.' Or whatever it is. So I think there's an incredible art. It comes back to your point about the wedding planner. You mention this a number of times, different kinds of intermediaries. You really have to be good at assessing the match--the quality of the match for both sides. You give the example in the book of the travel agent who says, you get to know the client well enough, you can say, 'I don't think that's going to be your favorite piece of that vacation. You probably would prefer to do this.' And I think similarly for these insulators, their job is to be tough and sometimes easy; but they have to know when; and they have to make sure that their client likes the stance they've taken. I think that must be very difficult. Guest: That's true. That's true. And I think you are getting into this other point about human intelligence. These are soft skills; they are sort of intangible. But when they work well, you don't necessarily see them working well, but there's a real art to a lot of what these people do.
56:12Russ: So, I want to close with an example close to both of our hearts, which is the publishing business. So, we live in a time where it's easy to publish your own book. There are many, many companies out there that will self-publish. The distribution channel is there: on Amazon they'll put your book up on their site; you can drive traffic to it. And you'll make "all the profit." And yet, you chose to publish your book with an intermediary. Which was an interesting decision you had to make. You may have used an agent. I did both for my last book. I gave away money to my agent and my publisher I could have kept for myself. And I think a lot of people when they hear how little authors make--they make somewhere usually between 10% and 15% of the book--they go, 'What? You provided all the content, and that's all you get?' And the answer is, 'Yeah; and it's a great deal. Because I have an alternative.' So it better be the case that they provide enough value to make that a decent proposition. Oh, and of that 15%, I give 15% of that to my agent. Or 10%--I don't even remember now. Sorry. But think how crazy that is. So, why do we do that? Guest: Yeah. Well, I would separate the agent and the publisher. But the publisher is a risk-bearer--it's the [?] example of a risk bearer. I get to keep all the profit, assuming there's any profit to be had, if I were to publish myself. Russ: Shhhh. Guest: With certain costs. We don't know if we'll ever turn a profit, and particularly in something as uncertain as the publishing industry. And any culture[?] industry--highly unpredictable. So, I think a lot of first-time authors or people who are not in the industry just don't appreciate how hard it is to sell books. The author really stacked against--the odds are really against you when you are publishing a book. Most books--I think the figure I've heard most often is 90% of books lose money. And the reason publishers are able to stay solvent is that the remaining 10% make so much that they cover all the others. And of course-- Russ: [?] don't know which they are in advance[?] Guest: Yeah. Yeah. So, as an author you are just not in as good a position to build a portfolio. You can not put out a hundred books and hope that 10 of them are going to make enough money to cover the others. Yeah. Russ: But the other factor is that the costs are there, that are just hard to see. It's not just that you might not get them back because you might not sell books, but you sometimes forget--there's editing, there's printing, there's the design of the cover; there's a thousand things that make a finished product attractive and that people find out about. And I think--it's just an example I like. Because there's so many times where people say, 'Oh, people are being ripped off,' and they don't realize what all the costs are. Guest: Exactly. Yes. And unfortunately the more--I don't want to say ignorant--the more new you are to it, the less you see those things. You may not even understand that you need a professional cover designer, because if your cover is not professionally designed then people won't even click on it, on Amazon, because it just signals right away that this is not a high quality product. So, you kind of have to be in the know to even know that you need a publisher.
59:59Russ: So let's close with a general question. We started off by saying that the Internet is going to kill off--people thought the Internet was going to kill off intermediaries. And of course it created through technology a whole bunch of intermediaries and we've talked about some of them--Kayak, Uber, Open Table. These are intermediaries where there aren't[?] people providing a skill. And yet there are still plenty of cases where there actually are human beings providing a skill, like the wedding planner we talked about, and many others. Where do you think it's going? And, what advice would you give to somebody who is worried about technology taking away their job? The travel agent of the past, a lot of them did lose their jobs. And I assume there are other jobs that are at risk right now because of technology and the ability to provide these services cheaper, of close enough value, that human intermediaries, human middlemen will be obsolete in certain areas. What are your thoughts on where that might be going and what advice you might give? Guest: Yeah. I think middlemen are becoming more important--as I hope I've made clear. I think the travel agent example is a good one. We have fewer travel agents; the ones who no longer provide that value, the mere order-takers, are the ones that have largely fallen by the wayside. Or if they exist at all, they are making close to minimum wage. The ones that are succeeding are ones who figure it out, what value they can still create in this day and age of abundant information. And in the case of a travel agent, it's a question of becoming an expert and being able to make sense of this glut of information that is out there. But there's certainly people who can be a certifier and use their expertise that way; who can commit to being in the business for the long term and also thereby be an enforcer; and all these other roles that I've talked about. So I think what the future holds in terms of human competition and competition from machines is to continually think about what you are doing to provide value and what you need to be charging that customers will be willing to pay as a fraction of the value that you are providing.

More EconTalk Episodes