Note Taking/writing

On Note Taking, Writing, and Competitive Advantage

Posted on by antonzitz
On Note Taking, Writing, and Competitive Advantage
On Note Taking, Writing, and Competitive Advantage

Do you want to think better? If so, writing is a good first step. A way to write better is to have a simple system that allows you to to take notes in a systematic manner, letting complexity bubble up from there (Beinhocker, 2006). This system will allow you to accumulate and compound insight, which Sonke Ahrens explains in his book: How to Take Smart Notes.

This system, known as a Zettelkasten, consists of two main components: literature notes and permanent notes. When you read something, have a pen in your hand to take notes on the content, summarizing what you think its important and what you don’t want to forget. You should then capture the bibliographic details in these literature notes so you have something to refer back to. Taking literature notes, in your own words, is a powerful test of concision and comprehension.

In the next step, you should make a permanent note, which is an elaboration or extension of your literature notes. When you do this, look at the literature notes you have just taken, and ask yourself, how can this idea contribute to different contexts (Ahrens 2017, 15), or why did I choose to write this down (84)? Other good prompts for developing permanent notes include: does this contradict, support, correct, or add to what I already have; can I combine two ideas to generate something novel; and what questions come from these ideas?

Over time, taking permanent notes is about making connections to other things you know, which comes from a mix of the theories, rules, mental models, and narratives you know today. Additional probing questions for permanent notes include: how does this fact fit into my idea of (x); can this be explained by theory; do these ideas complement or contradict one another; what is this argument similar to; have I heard this before; and what does (x) mean for (y) (Ahrens 2017, 60)?

When composing literature and permanent notes, I’ve found it helpful to use ideas from George Gopen’s Readers Expectations Approach, or REA, covered in his books: The Sense of Structure: Writing From The Reader’s Perspective & Expectations: Teaching Writing From the Reader’s Perspective. Your notes (both literature and permanent) should pass Gopen’s simple test of good writing: was the message delivered by the writer to the reader? If it was, then the writing was good; if it wasn’t, then the writing wasn’t good (Gopen 2014, 8).

Gopen’s focus on the stress position is also very useful for composing permanent notes as it helps clarify your thinking. He defines the stress position as the part of the sentence, usually at the end, that you most wanted to emphasize. For example, after reading the sentence you just composed, you may realize: there is more than one candidate for the stress position, implying you may be conflating or juxtaposing two separate thoughts; what you wanted to emphasize was not in the sentence, but in an implication hanging above it; or, you may realize nothing was worthy of emphasis, and as such, the sentence (or note) should be reduced to a phrase or clause (Gopen 2014, 14).

When storing a permanent note, don’t use topics and subtopics — you want to keep things as simple as possible and let complexity build from the bottom up. Identify notes in the context within which you may like to see them again, whether those words are in the note or not (Ahrens 2017, 38).

Why is any of this important? I believe both taking notes and good writing, enforces good thinking and can be a competitive advantage. As the world transitions to knowledge work, exercising this writing muscle is essential. Not only can writing confer an individual competitive advantage, but it can do the same in businesses.

For instance, in the book Working Backwards: Insights, Stories, and Secret from Inside Amazon, the authors explain how Amazon relies on the written word and have banned Powerpoint (Bryar & Carr, 2021). They point out that Powerpoint, due to its format, prevents ideas from connecting to one another and makes an audience more passive (83). Amazon instead has its teams draft a six page narrative, complemented by tables, graphics, and numbers, that is read at the beginning of meetings, which they believe allows them to think through issues more deeply, foster better discussions, and finally make better decisions (Bryar & Carr 2021, 83). As the work becomes more complicated and relies more on human capital as an intangible moat (Wu, 2021), writing is must.

Even when it develops new product, Amazon uses a document in the form of a press release and frequently asked questions, known as a PR/FAQ, to develop new products. This is a living document used to ensure a new product will benefit customers, determine whether its feasible, and whom you are reliant on to introduce the product.

In another example, Twilio, in its annual report, writes the following in the context of how it makes decisions: “Write It Down. Our business is complex, so take the time to express yourself in prose (emphasis mine)—for your sake and for the folks with whom you’re collaborating.”

In conclusion, if you want a leg up for yourself, develop a note taking system and take the pains to write clearly. Also, pay attention to organizations that prioritize writing — they may actually have an superior competitive advantage to those that don’t over the long term.

I would love feedback on this and any other examples of companies that focus on writing. Thank you.

Economic Development/economics

How Asia Works, A Summary Rephrasing

Posted on by antonzitz
How Asia Works, A Summary Rephrasing
How Asia Works, A Summary Rephrasing

I recently finished How Asia Works. Instead of my format I used in the last few post, I tried a slightly different format. I took what I thought were the most important ideas and wrote a brief essay describing the book. In other words, a brief rephrasing. Let me know what you think.

Introduction

How Asia Works: Success and Failure in the World’s Most Dynamic Region, a book by Joe Studwell analyzes the rapid economic development of Japan, Korea, Taiwan, and China. The author analyzes the successful policies of these nations and compares it to the less successful policies in Malaysia, Indonesia, and the Philippines. Studwell argues there is a consistent “recipe” for economic development: 1) developing agriculture as a form of “large-scale gardening” that maximizes output, putting a large and relatively unskilled population to work, 2) making investments in export-oriented manufacturing to encourage technical learning, and 3) developing a closely controlled financial framework in service of agriculture and manufacturing and not deregulating financial markets too soon. The interesting conclusion from this framework is the idea there are two types of economics: economics for early development and market economics for a developed country.

Step 1: Agriculture

A developing nation must first address its agriculture to put its population to work and maximize output. This takes the form of “large-scale gardening” which will not be as efficient as “scale” farming, but seeks to maximize output1. Economic theory tells you “scale” farming is the way to do this as its “efficient”. This may be true if your goal is efficiency, or to maximize returns on capital. In an early development stage, many nations should not focus on this, but rather on maximizing output. Many Northeast Asian nations that reorganized their agriculture did so by employing a land policy based on “land redistribution”.

Land redistribution was redistributing land into equal parcels for all as opposed to having concentrated land ownership. While it may seem drastic and unfair, it creates the conditions for perfect competition many economists dream of (Studwell, 2014, 10). This policy results in output growth as competition realigns and removes the perverse incentive of powerful landowners. For landlords, it is easier to raise their return by raising rent than by helping farmers increase yields. This creates a vicious cycle: higher rents result in less money for farmers to invest in the land, keeping yields low. Land redistribution policy breaks this cycle; research in the book explains equal land distribution correlates well with future economic growth (Studwell, 2014, 10). Once output is maximized, this prevents countries from having to import food and creates consumption for industrial development, leading to more political stability and upward mobility (Studwell, 2014, 66).

Step 2: Manufacturing

After developing its agriculture to maximize output, a developing nation should then invest in its manufacturing to promote industrial learning. The author points out these nations should begin with infant industry policy that protects domestic firms, allowing them to learn and grow. This policy highlights another interesting divergence between economic theory and economic history. While many economists will tell you to leave it to the “market”, history shows many economic powers developed with protectionist policies in early stages such as England, France, Germany, and the United States. After nurturing these domestic firms, you must allow them to compete, but not by picking winners or national champions. Instead, you eliminate these underperforming firms by shutting them down or forcing them to merge into more successful firms.

The next effective policy measure is something Studwell calls “export discipline”: the idea that as firms are learning, they should make goods for export to the international market as this is the true test of competitiveness, providing feedback on investments firms are making. To encourage industrial learning, successful nations such as Korea did not enter into joint ventures with multinationals as the arrangement creates technological dependence. Instead, Korea partnered with a Mitsubishi, a weaker Japanese firm, to learn its technology while also seeking out opportunities to learn about technology and processes from many other auto manufacturing firms. In addition, Korean firms did not want to blindly trust technology, but wanted to understand the “nuts and bolts” of how everything worked even if the process was outdated or more manual (Studwell, 2014, 142).

By understanding the basics, getting a variety of perspectives, and insisting on technological independence, Korea became a force in both automobile and steel manufacturing. To summarize, after a period of initial protectionism, a developing nation’s manufacturing policy should seek to spur domestic competitiveness and then international competitiveness via export discipline.

Financial Policy

In order to properly develop agriculture and promote industrial learning, a developing nation must tie its financial policy to these two goals. The simplest way to do this is through the banking system and provide credit based on export performance. Another means is to impose capital controls in early development stages. Some countries took the advice of developed nations that said to deregulate financial markets prematurely. This may help in developed countries but not in developing ones as capital markets are harder to control.

Premature financial market deregulation results in speculation and bubbles, not efficient capital allocation. A key takeaway from this book is that entrepreneurs are very similar while the policies surrounding them are different, yielding different results. This premature financial deregulation advice is another instance in which economic theory and policy do not align in an early stage of development.

Conclusion

In conclusion, economic development policy in early stages and economic theory do not seem to coincide judging by outcomes. In agriculture, a focus on output versus efficiency initially drives the gains to get a population working and productive. In manufacturing, protectionist policies to get businesses off the ground is effective based on history. The “market” cannot cure all: without land redistribution or incentives to export, landlords and entrepreneurs will take the “path of least resistance” to increase their returns, which do not align with national economic development goals in an early stage.

Incentives matter in all these stages and policy helps drive them. Finally, premature financial deregulation results in bubbles and speculation, not in efficient economic development or capital allocation. There are two types of economics: the stages described for development and a market economy based on efficiency, profits, and deregulation (Studwell, 2014, 269).

economics/investing/valuation

Expectations Investing, Some Interesting Takeaways

Posted on by antonzitz
expectations infrastructure
expectations infrastructure

Similar to what I did for Trade Wars Are Class Wars, I wanted to share a few takeaways from a book I recently reread: Expectations Investing: Reading Stock Prices for Better Returns by Michael Mauboussin and Alfred Rappaport.

I will provide a brief summary of the book and share some interesting takeaways I had from reading it. Later this week, I will post my notes from the book. While written in 2001, this book has a flexible framework that is still very relevant for business analysis. On a recent podcast, Mauboussin mentioned he may update the book. Fingers crossed!

Brief Summary

The book has a simple premise: stock prices reflect the sum of investor expectations at any point in time. To achieve superior returns, investors must calibrate and understand the expectations built into a stock price.

The authors advocate using a “reverse DCF” to determine expectations. In essence, a reverse DCF starts with the current price, backing into a set of cash flows that justify today’s price. You can start this forecast by using Wall Street consensus estimates and extend it out until you have enough years of free cash flow to get to today’s price. I would note the book has a website if you want to download the model and see it for yourself.

The book then provides a framework to analyze how a company creates value based on its value factors1. Next, the book goes into several frameworks to analyze both industry economics and competitive positioning such as Porter’s Five Forces, value chains, Clayton Christensen’s Disruptive Technologies framework, and others.

Finally, the book concludes with discussions of how corporate events such as mergers and acquisitions, share repurchases, and executive compensation can signal changes in expectations.

Mental Model for the Stock Market

This book dispels the notion that the stock market is “short-term” and explains that the market values stocks on long-term cash flow expectations. In 7 Powers, Hamilton Helmer says the same thing:

“In public markets, value is determined by and changes based on expectations of future free cash flows”. Just because prices move in the short-term doesn’t mean the market is short-term. This is just the market reacting to what current performance may indicate about future performance, continually reshaping future expectations.

The simple way to think about this is if the market is short-term, every stock would trade at 1x its earnings or cash flow, but they don’t. This implies the market takes a long-term view, but not necessarily the correct view. The difference between what happens and how the market views it shows where money can be made, a point the authors stress throughout the book.

The mental model I formed from reading this is the market values stock on long-term cash flow expectations, but allows investors to make short-term bets on long-term outcomes. The market provides liquidity so a short-term view can be expressed, but its still based on a long-term outcome.

Value Creation & Economics

I often hear words or phrases repeated so often they lose all meaning to me. For example, strategy used to be one of these words, but reading 7 Powers has help me clarify it in my mind a bit.

“Value Creation” is another one of these phrases. What does it mean? To me, creating value means creating benefits for the customer. However, any business can create benefits, but incurs costs to do so. In an economic context I define value creation as creating a positive spread between benefits and costs.

At a company level, it gets a little more specific. Is it enough to generate a positive spread between benefits and costs? The answer is no. If you borrow or raise money, investors expect to earn a return on that money you will be measured against. That is your opportunity cost of capital. If the return earned on your investments is less than that cost, you are not creating value.

Value Creation in the Company (ROIC)

In this book, the Mauboussin and Rapport explain that a company creates value when it invests at a rate that exceeds the cost of capital. How do you measure that rate? I believe return on invested capital (or “ROIC”) is a good place to start. So, just like above, you aren’t creating value in a business context unless you generate a positive spread between ROIC and the cost of capital.

To understand if a business is creating value you must understand its unit economics. What does this mean? I will let Mauboussin explain in a more recent piece he wrote as he says it much better than I can:

What is in an investor’s control is gaining a solid understanding of a company’s prospects for creating value. This requires a grasp of the basic unit of analysis, which answers the fundamental question of how a company makes money. The basic unit of analysis for Walmart, and other retailers, is the return on investment for a store. Net present value is the tried and true way to conduct this analysis. A store creates value if the present value of future free cash flows it generates exceeds the investment the company makes in it.

Michael Mauboussin, “One Job”

Studying unit economics helps you figure out the return on one store, customer, or transaction.

This book provides a framework to assess company prospects for value creation by identifying its value factors, which drive its value triggers and value drivers. Value factors are the underlying drivers of revenue, cost efficiencies, and investment efficiencies. These value factors determine the value triggers of sales, operating costs, and investments. In turn, these value triggers imply the value drivers of sales growth rate, operating margins, and the investment rate. These variables drive expectations of future free cash flows.

Real Options

Although the authors use a reverse DCF as the basis for investor expectations, they also acknowledge that not all value will be captured in a DCF. Sometimes, a company can have some of its value attributed to real options. Real options are business decisions to expand, defer, wait, or abandon a project which all have economic value.

The authors point out that you can analyze a company’s investing patterns to see if they exploit real options or contain some kind of embedded optionality. The book does a case study on real options using Amazon’s real option to further expand in the e-commerce market. When the book was published in 2001, Amazon had a market cap of $22 billion(!).

Amazon, the Real Options Machine

With the benefit of hindsight, Amazon has been a “real option machine” over the past twenty years. For instance, the birth of Amazon Web Services, reflects the idea of real options well. Amazon was having trouble scaling its IT resources quickly, so it organized its IT infrastructure into compute, storage, and database. It was quite good at it and invested heavily in it.

At some point, Amazon realized it could offer this business as a service, so its investments reflected embedded optionality to create a new line of business. Today, AWS may be the most valuable business Amazon has. They also seem to be in the early stages of doing the same thing with logistics($).

I don’t know if Amazon views it this way explicitly, but I am not sure they view costs as “costs”. What I mean is they look at large cost centers as an opportunity to offer services to others that will create not only offsetting revenue and but give them the ability to scale further, reducing units costs and enhancing scale advantages.

Inputs vs. Outputs

Mauboussin makes the point in this book and other places that multiples are not valuation. I’ve spent time in both investment banking and private equity, and more often that not, valuation is determined by some multiple range (usually chosen in reference to comps) applied to said metric. Slapping on a multiple on a metric confuses inputs with outputs.

What determines valuation multiples are the value drivers: sales growth, ROIC (driven by the operating margin and investment rate), and the discount rate.

Cash flows drive valuation and they can correlate to earnings, but are not the same thing. Earnings have a few weaknesses, with one of the most notable being they don’t account for opportunity costs nor investments in fixed and working capital.

Multiples in the context of M&A and Share Repurchases

Valuation multiples are not only bad for valuation work, but also for evaluating an acquisition or calculating the return on a share repurchase. Just because an acquisition is accretive based on multiples doesn’t mean it creates value. For an acquisition to create value, the present value of synergies must be greater than the premium paid. This makes sense: the present value of future benefits in a combination must be greater than the premium paid.

The same idea works for share repurchase. The return on a share repurchase is not the inverse of the P/E multiple. For example, if a stock has a P/E of 10, the return on a share repurchase is not 10%, or 1/10. The authors stress the return on a share repurchase must consider the cost of the capital and the discount at which the stock can be purchased relative to what management thinks its worth. For example, if a company has a cost of capital of 10% and stock that is trading for $8 a share, but management believes its worth $10, a share repurchase would have an an implied return of 12.5%2.

This framework is more robust as it implies there are diminishing returns to buying back stock. This would be true in cases where it trades above where management thinks its worth. Said another way, if management thinks its stock is overvalued it is probably the time to issue some!

Conclusion

I reread this book after reading it several years ago and came away with some rich notes I intend to post. Although its dated, a lot of the lessons are timeless and I suggest you pick it up.

Please provide any feedback below and I will update this post with a link once my notes are up.

economics/Global Trade

Trade Wars Are Class Wars, A Few Observations

Posted on by antonzitz
Trade Wars Are Class Wars, A Few Observations
Trade Wars Are Class Wars, A Few Observations

I just finished Trade Wars Are Class Wars by Matthew Klein and Michael Pettis, a book about the history of global trade that I think deserves your attention.

Below, I will provide a brief summary of the book along with some themes that I took away from it. If you are more interested, I will post more detailed notes next week and would also suggest picking up the book.

Brief Summary

The central theme of the book is that inequalities within countries are being misinterpreted as trade disputes between countries via “trade wars”. The current open system of global trade with the US dollar at its center was fine when the United States was large share of the global economy. However, the US economy is now a much smaller proportion of the world economy and the system is showing strains.

The book begins with a history of global trade starting with Adam Smith’s theory of specialization and ending with illustrating the complexity of Apple’s global supply chain. Next, the books goes into trade basics such as understanding that financial flows drive trade flows and how savings, investment, and trade imbalances occur. Next, the book provides historical context for two surplus countries (China and Germany) and a deficit country (United States). Finally, it makes some recommendations on how fix the problems of the imbalances that have occurred.

Next, I share some themes I came away with.

Global Trade Today

Before reading this book, I naively assumed low trade barriers were good for all. Unfortunately, that doesn’t seem to be the case. Open systems have benefits as well as costs. The open system of global trade today is subject to imbalances and bad incentives. These imbalances cause tensions between countries that really reflect inequality within countries that most can’t seem to square. Said another way, we like our cheap stuff, but it may be cheap because workers in other countries are not getting a fair shake.

For example, consider China and Germany, two economies that are export driven. They export more because their goods are cheaper. However, the authors argue this is not because they are more efficient, but rather the reflection of inequality within their countries. This inequality is a result of a domestic policy that subsidizes production at the expense of consumption. These countries suppress consumption of their workers to keep goods cheap. The value that doesn’t flow to workers instead flows to business owners who do not consume but rather save the capital they accumulate.

To compound matters, the wealthy people in these countries do not invest it domestically but abroad accumulating foreign exchange reserves. In effect, those countries overproduce for domestic needs, and export the difference to the global market. Even now, the two countries I mentioned above are now at odds ($) as they compete for export market dominance.

Unfortunately, these actions affect other countries as well. For example, from 2000-2010, the United States closed 66,000 manufacturing facilities as they are no longer competitive with global imports. Countries who force workers to under consume deindustrialize the countries that absorb the excess production.

Advances in technology have made global supply chains more efficient and complex, increasing profits, but also compromising both resilience and optionality. Make no mistake, optionality has value as we have seen during the COVID-19 pandemic. In an ironic twist this pursuit for efficiency has come at the cost of global supply chains being “weaponzied“, creating a modern day hold-up problem. Advanced economies always move up the chain into higher “value added” activities, but that doesn’t mean value chain activities such as manufacturing aren’t important, and in fact may be matters of national security.

Finally, this book pointed out that analyzing trade data and bilateral trade imbalances are not effective. Trade data is not useful because the country exporting the finished goods gets credit, but often raw materials and components originate in other countries. Bilateral trade data is also not useful given all the differences balance out on a global level. This means the imbalance between two countries is reflected somewhere else. Stated simply, viewing the US / China trade imbalance does not give you the whole story. Only global relations matter. This makes policy such as tariffs a tool that either has no impact, or the cause of additional harm.

Infrastructure & Innovation Drive Development

Initially, global trade was simple, consisting of raw materials and finished goods. There was really no infrastructure or ability to trade intermediate goods. The development of infrastructure, better communications technology, and the shipping container changed all of that. These advances spawned strong growth in global trade and led to complex global supply chains.

I thought I heard this story before, and I was right. The opening chapter of Economics of Strategy analyzes the business environments of the United States in 1840, 1910, and today to show that advances in infrastructure and technology reduced coordination costs. This allowed firms to develop from small businesses with a local presence to become large corporations who sought to expand markets and exploit scale and scope economies. The same thing has happened with global trade, but the markets are not going from local to national, but from national to global.

Capital Allocation

Capital allocation is a topic I often think about, but usually at a firm or even individual level. This book helped me realize capital allocation also occurs at the national level, which is really a sum of the individuals, business, and governments decisions.

At the national level, capital allocation is influenced by a country’s economic development priorities expressed through its monetary and fiscal policy choices. This book also highlights, much like people, nations are not always rational capital allocators. This of course makes perfect sense given countries are made up of people who make the decisions.

The authors point out that many countries accumulate US reserves not because they represent the best investments, but as a form of self-insurance against catastrophe. This is not to say it doesn’t make sense within the current system, but it is certainly not the most rational use of capital.

In other cases, ideology can trump logic and cause countries to misallocate capital. The authors point out that Germany likes a balanced budget and abhors public debt. They also contend Germany should have taken on public debt to improve its infrastructure instead of investing its excess savings abroad. This idea may have some merit.

Capital flows from other countries seeking safety have to go somewhere. If the government chooses not to issue debt, that capital makes its way to businesses or even individuals and this has been the cause of many recent bubbles, both historical and present day as the book explains.

Scarcity vs. Abundance

A theme I run into a lot is that of scarcity versus abundance. Ben Thompson argues in a world of scarcity, distribution power served as a moat for many companies that the internet has decimated. Now, value in the value chain has accrued toward the ends which he aptly explains how the internet is best defined by the smiling curve.

The authors make a similar point, but in terms of an age old tradeoff between savings and consumption. Delaying consumption to fund investment and increase production is not as important as it once was now that resources are abundant.

The takeaway here is more value will come from increasing consumption (demand) than by saving money that ultimately goes to find safety. The authors posit that consumption and production can push one another forward in the form of a positive feedback loop. Greater production, if transformed into higher wages for workers, would encourage them to consume more, which would help expand production. We often hear global growth is slow which makes this idea interesting to me. Is it because there is not demand, or that we are artificially suppressing demand through policy?

Conclusion

This book was an excellent primer on the system of global trade and provided great historical context that has led us up to today.

It also gave me a better understanding of how interconnected our world is and how choices made in one place can affect others far away. I am glad I read it as I feel we will hear a lot about global supply chains going forward.

Update: I posted my book notes here.

economics/technology

Cost Structure Impacts What Firms Do

Posted on by antonzitz
Cost Structure Impacts What Firms Do
Cost Structure Impacts What Firms Do

This seems obvious, but physical and digital goods have very different cost structures.

That said, it still warrants further examination and explanation for the layman. Today, I’d like to address distribution and production costs.

Distribution Costs

Distribution costs get your product to market. For physical goods, the marginal cost of selling one more unit are high. Physical products have higher marginal costs as they involve real world “frictions” such as shipping, storage, and retail / placement fees given the physical limits of warehouse and shelf space.

These costs can decline as a business gets larger. In other words, as you ship, store, and sell more products, costs should decline from Scale Economies. Eventually, these costs will reach some point at which they don’t decline anymore. This is know as diminishing marginal returns.

Now, consider digital goods. Examples of a digital good include an e-book, a software product, or a Netflix subscription. These goods have lower marginal costs of distribution. Typical distribution costs for digital products include website hosting fees and platform fees.

Website hosting fees are fixed and platform fees run about 30% of the retail price. Some think that 30% is high and that is why Epic Games has gone to war with Apple over this fee charged in the App Store. Ben Thompson has written a great article on Rethinking the App Store that I recommend you read.

The key here is marginal costs. Website hosting fees and platform fees can be very high in the aggregate, but low at the margin. In other words, variable distribution costs are higher for physical products than they are for digital products.

Production Costs

Like distribution costs, production costs differ for physical and digital goods. Physical goods are comprised of raw material inputs. To sell more products, you need more raw materials. This expense is variable. Variable costs change with output and fixed costs do not.

For physical products, the production process involves some mix of machines and people. These costs are actually semi-fixed in that they are fixed within ranges of output and variable between ranges of output. As a business grows its volumes, these costs will increase in steps. Just to recap, physical goods incur a mix of variable and semi-fixed costs to get the product through production.

Digital products do not incur variable production costs. They can be duplicated at essentially no marginal production cost. Digital products are developed by people such as software engineers or authors in the case of an e-book. There is a cost for sure, but this cost is fixed. The fixed costs of production may be higher for digital goods, but there is very little in the way of marginal costs.

Conclusion

Above, we have described how the cost structure of both distribution and production is different for digital goods. The composition of the cost structure has changed and resulted in much higher fixed costs and much lower variable costs.

Companies have more operating leverage than ever before. It is easier to get larger and there is more incentive to do so.

Bundling

This is how and why technology companies get so big. It is simply much easier to scale. Lower marginal costs have spawned bundling strategies and made them feasible and possible. Bundling allows companies to leverage the cost structure we allude to above of high fixed costs and low marginal costs. Bundling also allows companies to accommodate heterogeneous tastes and customer acquisition costs. Amazon Prime is a “super bundle” of physical and digital products.

The “Pie”

Economic policy decisions often face a trade-off between “enlarging or shrinking the pie” and “splitting the pie”. Technology companies face a similar trade-off with many decisions. Consider the Apple and Epic situation mentioned above. With low marginal costs, decisions to enlarge the pie will almost always be better than decisions about increasing your slice of it. Why? Because your dollars of profit will always be greater. The cost setup we mentioned above makes this trade-off easier than ever before.