Sometimes on Industry Focus, we dive right into a subject and forget to explain what some of the more in-the-weeds financial terms really mean.

This week's Tech show is a start at remedying that. In this episode, host Dylan Lewis and contributor Evan Niu explain exactly what they mean when they talk about the network effect, and then talk about how it affects companies like Facebook (NASDAQ:FB) and Snap (NYSE:SNAP); what operating leverage is, and how it affects companies; the difference between capitalizing and expensing, and what they mean for companies; what deferred revenue is, and why companies use it in their earnings reports; and more.

A full transcript follows the video.

This video was recorded on September 1, 2017.

Dylan Lewis: Welcome to Industry Focus, the podcast that dives into a different sector of the stock market every day. It's Friday, September 1, and we're going to try to liven up your Labor Day weekend drive with a little breakdown of some accounting terms. I'm your host, Dylan Lewis, and I'm joined on Skype by tech specialist and CFA charterholder Evan Niu. Evan, what's going on? Do you have anything fun lined up for the holiday weekend?

Evan Niu: Not really. The wife is out of town for work, so it's just me and the kids.

Lewis: Wild weekend with the kids! [laughs] I think you're still going to have a better weekend than me, though. I'll be moving this weekend.

Niu: We'll, that's not hard to be, anything is better than moving.

Lewis: We're actually taping today's episode Thursday morning, because I'm moving on Friday. Austin Morgan, our man behind the glass, is a new homeowner, and I think recently just went through a move.

Austin Morgan: I did.

Lewis: Do you got any moving tips for me?

Morgan: Don't do it! [laughs] 

Niu: Hire people to do it for you! [laughs] 

Morgan: Luckily my brother is a large human, he's 6'4.

Lewis: You're also a large human.

Morgan: I'm also a big dude, but my brother makes me feel small sometimes. Him and my cousin, I was like, "Hey, I'll give you $100 if you get all of these boxes and bring them over there!" And they did, and then I just had to unpack everything, which sucked.

Lewis: Man, I wish I could do that. I helped friends move this past weekend, and there were like 10 of us. It was amazing. In and out, it took no time. But, moving on a weekday means that you don't have that option.

Morgan: That's true. I was going to say, if you're helping friends move, then you have help, too.

Lewis: I've helped way too many people move to not have any help now.

Morgan: Yeah. I think, move what you can move and you need to move, and then Saturday, you're like, "Hey, what's up?"

Lewis: Call in the cavalry.

Morgan: "I got a case of beer, let's go."

Lewis: So, listeners, if your Labor Day traffic seems bad, just remember that I will be pouring with sweat and moving boxes on Friday and Saturday. To bring it back around to what we're going to be talking about today, Evan, I think every now and then on the show, we have a tendency to throw around industry terms, and perhaps we don't do the best job defining them. We get so excited about the topic we're talking about that sometimes we don't make it super accessible to folks who are either new to investing, or perhaps just starting out learning the tech space. So, I figure, today, we'll spend some time explaining a couple of those concepts that we might glaze over a little bit week to week.

Niu: Yeah. It's always good to give a little primer.

Lewis: I think one term in particular that we toss around a lot is the idea of the network effect. This is something that listeners and people who consume tech news probably see a lot, because one of the big tech names, Facebook, and then Twitter as well, Snap, kind of have this associated with them very frequently. And for those types of companies, when we hear network effect, we're talking about the direct network effect. The idea here is, an increase in usage of a good or service will lead to a direct increase in value for other users. And the layman way of saying that is, the more people who are part of something, the more valuable the experience becomes for the people who are using it, right?

Niu: Yeah. The classic example is always, the original phone. You would never buy a phone unless you could call someone with it.

Lewis: Yeah, a phone is kind of useless if you can only hold it.

Niu: Right. Of course, nowadays, we take it for granted, because everyone has a phone in their pocket. But the classic, Alexander Graham Bell from 50 years ago phone, when phones were a whole new thing, yeah, to get people to adopt that technology, it makes sense if more people have it.

Lewis: And looking at the social media companies, the example holds. The platforms become dramatically more valuable when you can use them to connect with people like your high school friends, cousins, aunts, uncles, etc. If no one is on Facebook, you really don't have much of an incentive to use Facebook. So, as people join it, it creates this virtuous cycle of other people joining. It's kind of a snowball effect. So, that's the direct network effect side.

There's also the lesser-known indirect network effect. The idea here is, increased usage of a product creates the production of increasingly valuable complementary goods, which then benefits the value of the original product. And that's kind of a complicated explanation of this term, and it might be easier to illustrate it with looking at Apple's consumer electronic devices like the iPhones, iPads, and then its Services segment, where things like the App Store are nested. So, you think about the consumer hardware side, and the larger that the installed base is of people with iPhones, iPads, things like that, the larger the app offerings are going to be -- because the iOS developer community will recognize the size of the opportunity -- the larger that library becomes, that increases the use cases and the functionality of the phone itself, which makes it more valuable to consumers that actually own the device.

Niu: Exactly. Importantly, I think, of that example, is that Apple isn't directly doing a whole lot, there. They're just connecting developers to consumers, and that indirect network effect is really robust and powerful. And because it's indirect, Apple doesn't have to do too much on its own other than operate the App Store and facilitate those relationships.

Lewis: And what we see with both examples here of the direct and indirect network effect is, once you get the ball rolling with these types of businesses, it can become a super powerful thing that just snowballs and really creates business pretty effortlessly for these companies, and winds up becoming a major catalyst for them.

Niu: I think one thing that's interesting when it comes to Twitter, since certainly all social media companies are very heavily reliant on network effects to really build their businesses, but I think one thing that's interesting with Twitter is, they don't enjoy that as much because it's such a public service. I think Twitter is more about following people you don't know, high-profile people, as opposed to your personal friends. Since Twitter is all public, you don't really need to join to derive value from the service if you just want to see what some high-profile people are saying, like politicians or celebrities or business people or whatever the case might be. So, it's interesting that by making their platform so public, it actually diminishes the value of network effects, because you don't have to join to actually be able to derive value out of it. Whereas Facebook is very much more about making personal connections with people you know, and you both have to join to get that experience.

Lewis: Right, because so many outlets are going to report on the relevant and newsworthy things that happen on Twitter anyways. And because tweets come in through search results, whereas you look at Facebook or Instagram, and very often you need to be a part of the service to consume any part of it, kind of the same thing for LinkedIn in a lot of ways. So, they don't enjoy it quite the same way. In ramping to the hundreds of millions of users they have now, it certainly helped. But it did, at a certain point, hinder their long-term growth. 

Niu: Right, definitely.

Lewis: I've done quite a bit of talking so far, so for our next term, I'm going to flip things over to you, Evan. Why don't we talk a little bit about operating leverage? This is something that's come up quite a bit in some of our recent shows.

Niu: For sure. It's also very widely applicable. Network effects is kind of limited to certain types of companies, but operating leverage is very universal. I think it's a very important concept, and [it] definitely deserves more of a deep dive.

If you look at a company's cost structure, if we go back to Econ 101 in college, if you remember, there are two types of costs that companies face. It's primarily fixed costs and variable costs. The combination, how those two types of costs combine, translates into your total costs. And how those costs spread out over your production gives you total average costs. And these variables change. If you look at, these factors are very important to how a company wants to scale. If we go back to a textbook example, it's like a t-shirt factory, the factory itself would be the fixed cost, but the cotton and all the inputs that you're using to make the t-shirts would be your variable costs. So, as you ramp up production, your fixed costs don't really change. You already built that factory, and it's there and it's paid for. Your variable cost is the inputs. The more shirts you make, the more cotton you're going to need. So, that proportion of how much of your cost structure is fixed versus variable goes a long way, because what it boils down to, you're able to grow margins and gross margins expand when revenue rises. So, you become more profitable as you grow bigger, which companies love. You can put up really strong results if you're really growing, and becoming more profitable while you do so.

Certainly, there's a lot of seasonality to operating leverage. But if you have a really high portion of fixed costs, and a relatively smaller proportion of variable costs, then when you're ramping up production, your total costs, once they're spread out over a greater base of units, your costs actually don't grow as quickly as your revenue, and that's how you get this really nice margin expansion. It's because you're able to spread out the fixed costs over a larger base load of units. So, that's really what it boils down to.

Lewis: And I think one of the most high-profile examples of this, that we've talked about on the show, is the difference between how Facebook has approached their business and their data centers and how Snap has decided to do it. This is something you've written about plenty for Do you want to dive into that a little bit?

Niu: It's a perfect example, because they're completely opposite ends of the spectrum in how they approach their cost structure. Facebook is much more the traditional, how most companies do it for a tech company. In the context of Facebook and Snap, one of the biggest things for these companies is how they deliver the service with their back-end infrastructure, or other people's infrastructure, in the case of Snap. Facebook has seven or eight gigantic data centers all around the world. Each data center is billions of dollars to build, and that's the fixed-cost part. You're putting billions and billions of dollars into this gigantic facility that enables your service. And that's a huge fixed cost up front. Then, the variable cost is the operating expenses associated with running those facilities. Think about electricity, utilities, the people that work there, employees, those are the variable costs. But, the point being, once you have that facility in place, and you're the one operating it, you own that facility, and as usage of the service increased, and certainly the way they monetize usage is obviously through ads, so once they start ramping up usage of the service and monetizing that usage with ad revenue while keeping their costs relatively fixed, because again, if you look at data centers, it's primarily a fixed cost with some variable costs in there, but it's very heavily concentrated toward the fixed side. The downside of having a really high fixed cost base is, if things go down and the business starts to deteriorate, now you're on the hook for a really big cost, and you have no way to pay for it. A good example there is the auto industry. The auto industry has a huge, very high fixed cost basis because they operate these gigantic factories. That's why they need to be always constantly selling cars, and they get pinched really hard when there's an industry downturn.

Now, if we flip that around to a company like Snap, their strategy is the complete opposite. They completely outsource all infrastructure to larger companies that are very good at this stuff, Google Cloud and Amazon Web Services. So, they don't own any of their own infrastructure. So, in a sense, they have no fixed costs related to their hosting strategy. It's all variable costs. So, every time usage increases on your service, you're paying increasing fees, so your costs scale up almost as quickly as your revenue scales up. Whereas, for Facebook, your total costs don't scale up, because you're increasing usage. So, that's why we've said before that it's going to be extremely hard for Snap to put up any type of operating leverage, because usage of the service directly translates into commensurate increase in variable costs. And it's true that they've been able to get some better pricing negotiations, so that's helping them. That's why they're able to have some improvements compared to a year ago. A year ago, hosting costs would exceed revenue, and now it's getting more in line to where they can actually turn a gross profit. But, that's not to say it's impossible. And, again, there's other factors, because they're negotiating pricing. But, fundamentally, it's harder compared to if you were to just do it yourself.

Lewis: Yeah, it's a lot tougher for them to wind up doing anything on the margin side to really expand. And this might be a conversation that our listeners might already be kind of familiar with. This is something we touched on a couple times specifically with some of these businesses. But, I did think it was worth diving deep into. We have a couple more topics that we want to cover, and I swear we will use different companies in our examples.

But, before we get over to the back half of the show, I just want to take a moment to talk about The Motley Fool Investment Guide. This book is the succinct summary of our investing philosophy. It's a playbook for everything we do here at the Fool, and a great guide for long-term investing. The third edition is a massive update on the second edition, even has a new section on options. Check it out at You can pre-order your copy before it goes on sale September 5. Again, that's

Evan, I think we both got a little rant-y in the first half of this show. [laughs] We both kind of got really excited about the topics we were talking about, and did not dialogue too well. We'll try to remedy that here in the second half. One of the other things we want to discuss was the notion of capitalizing versus expensing purchases. This is a bust-out-the-accounting-books type discussion. You're the one with the CFA designation, here. I'm going to let you start to explain, and I might ask some dumb questions here and there to help clarify.

Niu: Sure. So, this is a pretty big one, capitalizing versus expensing. Let's say you're a business, and you incur a big cost for something. Let's say you're building a factory for $100 million. If you're building a factory, that's a long-lived asset that you'll be able to use for some period of time after that. Five, seven, 10 years later, you could still be using this factory. So, if you're putting investing money into a capital asset like that, you would do what's called capitalized purchase, or capitalize the expense. Whereas expensing goes on the income statement and directly comes out of revenue, it affects your reported net income, and it's directly on the income statement, which is one of the most important financial statements that investors look at when they're looking at, what's your net profit, your net margin, your earnings per share, etc. So, if you were to expense the entire $100 million of that factory immediately, you're taking an immediate $100 million hit on your profitability.

Whereas, if you capitalize that asset, you don't put it on the income statement, but you say, now, we've built this $100 million factory, you put a $100 million asset on your balance sheet, and then you depreciate it slowly over time, over seven years or whatever the useful life of this factory is. So, instead of booking $100 million up front, instead you just book it as an asset and depreciate it slowly over time. So, in essence, it's a way to spread out that cost over time. It's a really important distinction of one that is or is not appropriate, because typically, with capital expenditures, those are all capitalized. If you're building big investments, those depreciate over time. Whereas things like operating expenses that are like regular expenses, that's something where you definitely need to be expensing it as opposed to capitalizing it.

Lewis: This is a concept that, to me, at a glance, seems like a very in-the-weeds accounting discussion. But when you start to work through it, it makes a lot of sense. It just makes common sense that, if you're going to use something over an extended period of time, why would you only recognize the cost of it immediately? Why wouldn't you do it over several years, right? 

Niu: Exactly. The sniff test is, if you're building something that you can use, then yeah, it makes sense to capitalize. But if you're just incurring an expense for your monthly utility bill, you're just using something, you don't have something left to show for it, you're not building an asset. Or, another example would be your payroll when you pay your employees, that's obviously something you expense. It's not something you can put on the balance sheet. And the most famous example of this, where there is some potential for accounting shenanigans, WorldCom was the biggest case of fraud around capitalizing versus expensing in the early 2000s, a year after Enron goes bust, then WorldCom --

Lewis: Not a great period for the accounting industry.

Niu: [laughs] Right. So, WorldCom was basically inappropriately capitalizing things as opposed to expensing things, because this is the heart of their scandal. They were taking operating expenses, like their payments to enable some of the services from their partners, etc. But instead of recognizing all these costs upfront, they were improperly capitalizing these costs, which again, allows you to spread out the costs over time. In doing so, they minimized their expenses in the current near-term period, which then goes on to inflate your profits because you're not fully recognizing your actual expenses, and you're basically mistreating these expenses because when you should be expensing them, you're capitalizing them and putting them on your balance sheet, and then you depreciate them slowly over many, many years. So, you're only really seeing a tiny portion of that upfront. It was something like $4 billion in fake profits that they were inflating, I don't remember the exact figure. But, the point is, at the heart of that scandal, they were improperly capitalizing expenses that should have been expensed in the current period, and they were doing it to an insane degree, they were going completely wild with it. And, of course, it lets them put up these really strong profit numbers, which investors love. But, it turns out, these numbers are all fake.

Lewis: So, I think, the shorthand for investors who are just beginning to understand this topic is, the idea of a business capitalizing is great, so long as it is an asset that has a long-term, useful life, and is not the operating expense that you're talking about here.

Niu: Right, there are definitely times when it's very appropriate -- for capital expenditures in particular. After WorldCom, I don't think a lot of companies are doing shady things with this kind of accounting. But it's something that's always good to be aware of, because the potential is there for abuse, and it generally gives investors a better insight into how and when and where these costs are recognized, whether or not they're put on the balance sheet or the income statement.

Lewis: For the last thing that we're going to hit in this discussion, I think we're going to flip chapters in the accounting textbook and look over at deferred revenue. I think, maybe one of the better ways to illustrate this is to go through an example. Say you're a business, you have a customer make an order and pay you upfront. But, the good or service that you're responsible for is not immediately delivered. So, it isn't fair for you to then immediately recognize all of that revenue, because you haven't necessarily incurred the costs of fulfilling that order yet. So, taking all the revenue at once would throw your numbers out of whack. There would be a lot of lumpiness in your revenue recognition. But, you have the cash. They did pay it, it's seeming there on your balance sheet. And they need something on the other side of the balance sheet to balance out. So, generally, what a business will do is record that advance payment as a deferred revenue liability on the balance sheet, because the assets and liabilities, equities portions of the balance sheet need to be tight. It is considered a liability because the company is on the hook for fulfilling this order. Again, I will gut check my explanation here with the CFA charterholder. How does that sound, Evan?

Niu: Yeah, that's exactly it. A good example is, imagine a magazine subscription. If you pay $120 for a magazine subscription for the whole year and get one issue a month, the company that's delivering that only gets to recognize $10 a month, but they've already collected $120 from you. So, they would initially record $120 of deferred revenue. Every month they deliver a magazine issue to you, they recognize $10. They already have the cash, as you mentioned. It's already recorded on the asset side in the cash line item. But then, over time, as they recognize that revenue, it switches from a liability to the equity portion, because then you're earning it, and it's actually yours now.

Lewis: And much like our capitalizing expenses discussion, this sounds like scary accounting, but the reality is, it's kind of common sense. If you're delivering something monthly to someone over the course of a year, it would be silly to recognize all of the revenue in the first quarter. It just wouldn't set you up to have nice, stable, consistent financials over the course of the entire year.

Niu: Right. We see this a lot with any company that has a subscription model where they bill in advance for, let's say, a year. Obviously, if you have a service like Netflix (NASDAQ: NFLX), where people are billing month to month, you don't really get a big balance of deferred revenue. But, if you're billing for a full year in advance, particularly for enterprise software companies, which we see a lot of, then you definitely going to have a very large portion of deferred revenue in there, because they haven't delivered that service for their balance of the year.

Another time you'll see this is where companies like Apple, for example, give you software updates for free over the life of the device. So, even though they're selling you an iPhone right now, they're promising to give you software updates for free for years after that. From an accounting perspective, those software updates have value. So, what they do is, they break out a piece of that sales price and assign it to software updates that they're going to give you later on, and because they deliver those over time, then they recognize that revenue. When you pay $700 for an iPhone, I want to say it's $29, something around $30 of that, that's set aside as deferred revenue for things they're going to give you later on. You'll see this sometimes with hardware products, because if there's an ongoing commitment to deliver things after you purchase, then there can also still be deferred revenue implications.

Lewis: You know, Evan, you just taught me something. I didn't know that. [laughs] For as much as we talk about Apple -- and I love doing the quarterly earnings show with you -- I did not know that about their business.

Niu: That's the bulk of where their deferred revenue comes from. Certainly, Apple doesn't have a whole lot of subscription Services. I mean, they certainly do, but that's not where that deferred revenue balance is coming from, if you look at Apple's balance sheet, just as one example.

Lewis: Evan, I see Chris Hill in the studio getting ready to take Market Foolery, so I think we're going to have to wrap, here. I hope you have a great holiday weekend with the kids.

Niu: Yeah, have fun moving! [laughs] 

Lewis: [laughs] Yeah, it's going to be great! Before we close out, I want to give listeners a heads up about the Industry Focus holiday schedule. We're going to have a bonus episode of IF posting on Saturday morning. We're calling it Industry Focus 101. And we actually got all of the hosts in the room, including new energy host Sarah Priestley, to chat about some of the big trends and stocks we're watching for the rest of the year. We had a really good time making it, so I hope you guys enjoy it, and I hope it helps you handle us not putting out a new show on Monday because of the holiday. Listeners, of course, if you have feedback or questions, we would love to hear about what you think about that episode in particular. Shoot us an email at, or you can tweet us @MFIndustryFocus. If you're looking for more of our stuff, you can subscribe on iTunes or check out the Fool's family of shows over at 

As always, people on the program may own companies discussed on the show, and The Motley Fool may have formal recommendations for or against stocks mentioned, so don't buy or sell anything based solely on what you hear. Big ups to Austin Morgan for all his work behind the glass. For Evan Niu, I'm Dylan Lewis. Thanks for listening and Fool on!

This article represents the opinion of the writer, who may disagree with the “official” recommendation position of a Motley Fool premium advisory service. We’re motley! Questioning an investing thesis -- even one of our own -- helps us all think critically about investing and make decisions that help us become smarter, happier, and richer.