Replacement Cycles: Malfunction & Innovation

In analyzing the sources of supply & demand for an industry one quickly becomes accustomed to the idea that demand moves in cycles. Economic cycles pervade mainstream economic thinking, so it would be natural to expect demand fluctuating with economic environments. As a result either supply or price also moves in cycles. If supply was fixed and demand cyclical, prices would be cyclical. More realistically, supply reacts to realistic expectations of future demand and prices react upwards, but not as intensely.

In most industries, the primary determinant of cyclicality is the volume of goods sold. Thus, an industry might be called “peak-cycle” if the volume of goods demanded are in excess of normalized expectations.

However there also exists other cycles of demand that impact the volume of goods demanded and their respective prices. One of these alternative impacting cycles is “the replacement cycle” which drives demand through a renewal of parts & products.

The simplest example of a replacement cycle is your consumption of water-heaters. Every time you buy a water-heater, you can expect it to last 8-12 years depending on price. As soon as your water-heater breaks, you either take cold showers or buy a new one. Your individual demand for water-heaters moves in a replacement cycle.

Replacement cycles are driven by obsolescence of products. There can be multiple reasons for obsolescence, each with a different cyclical effect. One reason is product-malfunction. I’ve titled these malfunction-cycles. The water-heater is an example of an individual malfunction-cycle of demand. Another reason can be that the product is now vastly inferior to other goods, leading to a required upgrade. I’ve titled these innovation-cycles, or upgrade-cycles.

Both are examples of recurring demand, which is an extremely positive attribute for any market as it lowers uncertainty, leading to lower borrowing costs and higher stability of income. The effects of each cycle, however, differs vastly based on a series of attributes. Each cycle must be analyzed independently.

The Cycle of Malfunction: Economic Cycles, Smoothing, and Catalysts for Bunching.

The cycle of malfunction on an aggregate level is not necessarily cyclical, despite its name. If the purchase of a new water heater was entirely independent of any other variable, water heater replacements would not show cyclicality. Why not?

Seeing as the purchases were not lumped together, but spread out evenly over time, the replacements also occur randomly across time in non-clustered ways. An example could be an extremely short replacement cycle good (though not malfunction-based) is toothpaste. As users generally buy the same amount of toothpaste independent of economic environment, demographics, weather, or most other things – the replacement cycle is flat in nature, leading to highly sustainable income and a lack of cyclicality. This has a smoothing effect on the economic cycle. While toothpaste is not a malfunction-based good, the lesson is that short replacement cycles and random purchasing can smooth out cyclical effects.

Yet most malfunction-goods have inherent cyclicality. Water-heaters are usually bought during housing booms and are also dependent on demographic patterns in home-buying ages. Water-heater demand therefore generally spikes 8-12 years post housing booms. Luckily a series of housing booms, over time, will create something akin to smoothing, as the demand from each boom follows the next.

Importantly, the length of the replacement cycle results in the recurring demand “popping up” at non-cyclical peaks in the next cycle. The cyclicality of malfunctioning-goods with relation to the economic environment is not necessarily procyclical, but often just cyclical independently of current economic conditions.

Malfunction demand is either smooth across time, or has cyclicality depending on the previous economic cycle. The demand can also be related to the current cycle, if replacement of the product can be temporarily postponed for economic reasons.

Sometimes malfunction-based demand have a catalyst for turning cyclical. Fireman-equipment, such as oxygen tanks, are largely a malfunction-good that generally last 15 years but often require premature replacement on a legislative basis. Except for the occasional budget difficulty, demand for oxygen tanks had an extremely non-cyclical tendency. Recurring demand was smooth across many years as each individual firestation had to replace their gear throughout the year, with no relation to the replacement of other fire-stations.

Then the 11th of September attack in 2001 occurred. Post-attack, George W. Bush Jr. made massive budget allocations to renewing all fireman-gear. This has created a clustered replacement-cycle, as all replacement demand now occurs around the same time, increasing cyclicality in the industry of supplying safety gear to firemen.

To conclude: Malfunctioning goods can lead to highly dependable buying patterns. Sometimes the buying will be constant, sometimes it will follow the last economic cycle, the current one, or be entirely independent of economic cycles as a result of historical events.

When buying companies in an industry with goods that are a “must-have” and last long periods of time, consider how the replacement cycle could look based on common-sense reasoning.

The Cycle of Innovation: Tech-Tock, Tech-Tock.

Some industries have demand for replacing products that aren’t even broken. Crazy, right? Not really.

Many can remember their kids begging for an Iphone instead of a flip-phone, products become obsolete all the time without breaking. Innovation can drive efficiency. With large enough efficiency-gains, consumers and businesses alike will overcome sunk-cost thinking and upgrade.

The presence of an innovation-cycle is mostly apparent in industrial components, not consumer-goods. Consumers will not go out of existence if “operating inefficiently”, but many companies in competitive industries will. Thus a product innovation can lead to forced-upgrading if it’s markedly superior to prior products.

Oftentimes there will be one critical component that forces the rest of the adjacent industries into upgrading. Thus the demand for the producing industry is highly linked to the innovation-cycle. If the cycle of innovation is predictable, companies will even stop replacing old products as they await the newest innovation, leading to a “double-effect” of decreasing demand prior to the innovation and catapulting it post-launch.

The easiest example of innovation-cycles/upgrade-cycles would be the PC upgrade cycle, primarily driven by Intel CPU-releases. The latest was Skylake from Intel. The cycle generally renews every 3-4 years (though wise men expect it to begin growing longer, perhaps 5-6 years). At the end of the cycle companies will be reluctant to expand, thus a changing cycle-length throws considerable uncertainty into the mix.

Cycles often have intra-cycle signs that indicate progress, such as Intel’s previous tradition of first releasing a new microarchitecture, then a new process. Thus, a new microarchitecture was unlikely to be released prior to the new process. There is usually plenty of industry research on a specific innovation-cycle.

When looking at bonds & equities for companies that operate in an industry known for product innovations, especially commercially, approximate the position in the upgrade cycle using “tell-tale signs”.

Conclusion & Takeaways

We’ve hopefully demonstrated that demand can have multiple overlapping cyclical aspects that influence each other. These cycles are important adjustments for considering the price of a business when investing.

In general, recurring demand is an extremely attractive feature, but it heavily affects which attributes dominate in the industry.

In industries with innovation cycles, speed is the name of the game. If you’re the fastest at implementing and getting new innovations out, or the fastest at adapting to new product cycles, you generally win. This promotes engineering talent and R&D as a prime determinant of success.

If your business is based on malfunction, then the speed and ease of replacing the product is essential. Often brands and distribution channels dominate as determinants of success. Thus the marketing and sales organization should be a core focus.

I hope the article was informational. I hope you payment will be sharing the article or commenting with constructive criticism.

Good luck, and good investing.

Matthias Herskind.


Exclusive Markets & Economics of Density

Landline telecommunications […] involve significant fixed costs within each regional market, which are a requirement for economies of scale. These economies have created barriers to entry, protecting the incumbents. Potential entrants would have to seize sufficient local market share to become viable competitors, and the incumbents’ existing degree of customer captivity has made this difficult to achieve.

By contrast, global markets for long-distance telecommunications, film production, recorded music, and books are so large that they will support many entrants, each with a relatively limited market share.

– Bruce Greenwald, All Strategy Is Local.

This essay is one of the intermediate level steps to understand the waste industry as outlined in our primer on the waste industry.  Of the many peculiarities in the waste industry, economics of density & exclusive markets are the most important industry dynamics to comprehend for competent analysis of the waste space.

This essay aims to explain the distinction in market types that waste operators participate in. There have been a multitude of studies on “how waste markets function”, but geographical areas differ widely in trash collection. Context is king when it comes to analysis.

Classification: Natural Monopoly

The waste industry is defined as a natural monopoly by the United Nations Environmental committee.  Below is an excerpt from the 19th page of a UN committee discussing the waste industry:



So what is a natural monopoly? A natural monopoly is when economies of density combined with high asset depreciation ensure that full volume with one operator provides the most efficient solution for stakeholders.

Let me unpack that sentence in layman terms:

  • Economics of density is when a business can provide the same services with the resources by having more customers in one specific geography. Imagine that a neighborhood of 5000 people need to employ guards to patrol their street at night to dissuade criminals. Here are two scenarios that illustrate how economics of density work.

    Scenario 1: Each household uses a different residential guard company. In front of each house, a guard is stationed (5000 guards). Criminals are extremely dissuaded.
    Scenario 2: The entire neighborhood employs a single guard company. The guard company sends out 50 people on patrol each night. Criminals are extremely dissuaded.

    Scenario 3: The entire neighborhood employs a single guard company. The guard company sends out 4 people on patrol each night. Criminals are moderately dissuaded.

    Sending out 50 people to patrol a tiny neighborhood is still extremely excessive, but I wanted to emulate the effect of scenario 1 at a 1/100th of the cost. It should be clear when comparing all examples that a single provider uses the least amount of resources to achieve the desired effect.

  • Asset depreciation is when a business needs to replace expensive equipment to continue operating. If you run a personal taxi service (perhaps through Uber) then you need to replace your car every 6 years. So while each tour might only cost new fuel and personal wages, you still need to earn enough excess profit to cover replacements later on. This is why depreciation is listed on the income statement.In very competitive industries companies with large asset depreciation can often be long-term unprofitable as they are forced to sell their services above immediate cost, but below costs when including the need for replacement.

By having these two attributes the waste sector is much more stable and efficient through communal contracts rather than individual use.

As a result natural monopolies often consolidate customers and bid them in contracts, to ensure that a bidder can be insured optimal volume while still promoting price competition through the auction process. 

To provide a concrete example for economics of density in the waste sector: Imagine 6 areas that need trash collection.


Authors creation. Simplified Model, Economics of Density.

If a collection route has to operate above economic cost (including depreciation of trucks) there are two important factors:

1. How many circles you operate in (keeping utilization high on fixed assets)

2. Keeping the length of lines to a minimum (keeping variable costs low)

Below are two images that contrast potential solutions.


Authors creation.

It should be clear why the entire waste industry operates on contracts. This is one reason why waste operators generally consolidate areas over time, and the reason that operators are generally densely focused in terms of geography. Only the big players have nation-wide operations.

There is an important caveat to above model that illustrates the first major industry dynamic. Only certain markets have economics of density. Whether a market benefits from economics of density depends on the volume in each market. 

Volume & Economics of Density Explained

There is only a certain amount of space in a garbage truck. If a truck could be filled from one circle there would be no regional economies of scale.

In Bruce Greenwalds book Competition Demystified he outlines how a supermarket in a rural town will have complete dominance. There is simply not enough demand to sustain two supermarkets, so nobody attempts to enter. However a metropolis will have enough volume to sustain multiple supermarkets who will have to compete on prices.

The same is true in the waste market. De-minimis size requirements make isolated markets attractive. If (in our illustrated example) each circle had enough trash to sustain a whole operation, there would be no economics of density. As such there would be regular price competition.

When operators who focus on exclusive markets, such as Waste Connections (NYSE:WCN), acquire urban contracts the EBITDA-margin is at least 500 basis points lower. 

The team at Waste Connections generally include the following slide in their presentations.

Source: Page 5.

We cover integrated versus non-integrated in our article on: “How Landfills Drive Consolidation”

The table clearly shows that exclusive markets are preferable to competitive markets. As explained above “exclusive markets” simply means that the areas are rural enough that regional economies of scale manifest themselves.

What Does This Mean for Investors & Entrepreneurs?

For investors it is important to recognize that margins are not comparable across companies with different market compositions. In the United States operators such as Waste Connections have margins in vast excess of peers such as Republic Services or Waste Management, but not because the other two have inefficient management. One cannot simply expect margin differences to close or use them as arguments for return to the mean.

Investors should also consider if rural markets will turn urban or enough people will migrate to make rural markets less attractive over time. Investors should keep market-composition in mind when setting margin-targets for companies that haven’t realized full utilization on assets.

For entrepreneurs the exclusive/non-exclusive market paradigm illustrates what both companies and customers are interested in. Waste operators are interested in purchasing solutions that solve volume-issues in urban areas.

Customers, such as municipalities, are interested in making bids competitive in rural areas. The task is vastly more difficult for an innovator here, but could include alternative logistics solutions or software that optimizes consolidation of municipalities for contract-offerings.

Most of all both investors and entrepreneurs need to comprehend what drives profitability and market paradigm for businesses and customers alike.


How Accounting Distorts The Staffing Industry

And he read Principles of Accounting all morning, but just to make it interesting, he put lots of dragons in it. ― Terry Pratchett, Wintersmith

Accounting is a wonderful tool for understanding a business. It outlines the amount of capital employed, how the capital is financed, the amount of cash the business is producing, and illustrates expenses across all levels of activity.

Yet accounting can also be treacherous. Revenue can be front-loaded, expenses can be capitalized, and a thousand other tricks can lead to a mirage for investors.  There are multiple tools to counteract these mirages. Cashflow awareness, footnote diligence, and a broad knowledge of accounting all reduce the risk of miscomprehension.

Like any tool accounting is great if you have control, but a lack of comprehension or focus leads to danger.

In the staffing industry the unique accounting related to sales & profitability lead to a wide range of misconceptions both when screening for the stocks and when judging investment merits. I’ve heard multiple times that staffing had too low margins to be an investable business. I’ve also heard that the Price-to-Sales ratio indicated a good purchase on many staffing stocks.

Below is a small compilation of times that not knowing the accounting on staffing revenue would have led to overlooking or overestimating staffing stocks.


Produced by Author. Sources range across 5-10 investor-focused articles from major websites such as Forbes, Nasdaq, etc. 

The Business and Accounting of Staffing Agencies

The staffing business is at its core a brokerage business. As covered in my primer the business centers on joining individuals seeking temporary placement and corporations seeking a temporary workforce for several reasons.

Regular brokerage businesses carry transaction volume as a separate footnote or line item and the take-rate is revenue. If you want a boat and I introduce you to my friend who owns one in exchange for a $200 commission, brokerage accounting would have $200 as sales while traditional retailers would carry the whole boat as sales (seeing as they carried it at cost in inventory).

Visa and Mastercard don’t carry total transaction volume (in the trillions) on the income statement, only their >0.5% take. Online brokerages such as Ebay carry their “bite of the cake” as net revenues and show gross merchandise value in footnotes. Actual stock brokerages use the same model.

For sales almost every brokerage model doesn’t use the gross transaction volume, but only the slice that ends up at the firm. Investors could be excused for assuming that staffing would operate similarly, but it doesn’t.

Staffing sales are counted as the total wages paid to the employee. That includes employee wages, workers’ comp, federal unemployment, state unemployment, health, and insurance costs. Gross profit is the take-rate of the staffing agency. The categorization of payments leads to the what should be sales ending up as gross profit.

The sales of staffing agencies are in effect the GMV of online retailers, the transaction volume of card networks, or the total sums managed by brokerages. The gross profit constitutes the true sales.

How the Accounting Distorts Screening and Cross-industry comparisons.

As the sales of staffing agencies include ALL the wages paid to their placed temp. staff, the margins are extremely thin and the price-to-sales ratio is equally low.

For investors focused of net profit margins (as demonstrated in the top image) this might lead to categorizing staffing firms as ‘dogs’ and commoditized. Many investors have minimum requirements for margins which is extremely rational. It is traditional micro-economic theory and investment considerations (from the time of Graham) that the lowest cost manufacturers (lowest margins) are leveraged towards commodity prices.

It is entirely reasonable to avoid certain commoditized, low-margin business models, but not when they’re only optically low-margin due to an accounting exception.

For investors focused on low price-to-sales (as showcased in the photo above) the unnatural accreditation of sales will result in a seemingly “value-style” stock on a revenue basis.

The low margins might inspire false confidence as I’ve heard from some commentators on manned security and staffing agencies. They will argue that margins are so low that nobody wants to disrupt or compete. In reality the margins at some firms are in line with most industries today on an adjusted basis.

On the other hand I’ve seen arguments that certain aspects of staffing that doesn’t involve employee wages, such as recruitment process outsourcing, will face contracting margins because they’re so “attractive” when compared to staffing revenues.

All of these arguments are entirely invalid based on traditional accounting measures. My argument isn’t that staffing stocks are likely to be under- or overvalued as a result of this phenomenon.

My comment is simple: Be extremely wary if you see an argument related to margins or price-to-sales of staffing stocks.

Staffing Companies on an Adjusted Basis

I’ve tried to summarize the difference in a comparison between a “normalized” income statement and the status quo income statement.


(1) An adjusted Income Statement contrasted with the current version.

While the absolute numbers are the same (e.g. both versions would end up at the same absolute profit figures) the inclusion of wages in sales and cost-of-goods sold severely disrupts any investor screen aiming at “high-margin” companies.

Further Complications

So you’ve learnt how staffing accounting differs widely from what is usually applied to brokerage-type businesses. That means you’re ready to go screen for the best staffing companies.

Not so quick young buck.

There is yet another complication. Staffing companies often derive income from permanent placement fees/direct hire fees. Direct hire fees are charges to staffing agencies for finding an employable candidate.


An explanation of Direct Hire Fees from Talent Plus Staffing.

Direct hire fees don’t have the wage component in goods of goods sold. Aforementioned means it’s extremely high gross margin as most of the expenditures are based in operating expenses.

Direct hire fees are good examples of products that should be treated as direct revenue. As the two different fees mix, it skews the gross-margin upwards. Be aware of this fact when comparing margins across companies within the industry. Industry-relative comparisons are disrupted by direct hire fees not being treated like pure staffing revenues.

With that last explanation in mind you’re now cognizant of the intricacies surrounding staffing margins. Go and impress some girls at the bar with it!