Codex <> Tactics <> Gallery <> Allies and Enemies <> Tales of the Tigers
Economics of Warhammer
Robert Van Pelt
“I am selling my army and quitting the game.”
Do phrases like that sound familiar? I know I have heard them on many occasions, both from people at my local game store and on the Internet. Have you ever felt that way? The thought has crossed my mind at least once.
Every so often, GW hikes up the prices on some or all of their models and then that is followed by the laments of those who feel that the prices are too high and they are being ripped off. But is GW justified in their price hikes?
Throughout this article I hope to find some answers to these questions. Mr. Kilgore, the owner of this site, has addressed some of these concerns. Others have offered good advice at how to make the hobby less of burden on the pocket (Check out Patrick Eibel’s Thousand Points of Light series and his article on plastic miniatures).
I hope to tackle this issue in a somewhat different way. I intend to take a more in-depth look at exactly how much it costs to produce a model and how GW prices stack up in the larger world of economics.
Who am I to attempt such a feat? By night, I make my rounds among the gaming community playing RPG’s, miniature war games, MMORPG’s, and first-person shooters, along with the occasional board game. But obviously that isn’t my day job. I am a graduate from the University of Toledo in Mechanical Engineering and have hands-on, practical experience with various manufacturing processes, including injection molding. Another facet to the engineering profession is heavily involved in economics. Engineers are responsible for everything from planning budgets, to sourcing suppliers, and performing quality analyses. Ultimately, this article will take the form of a small project analysis, studying the costs of production as well as the costs further along the logistics line of the product.
As I start, let me be very upfront in saying that throughout this article, I have made some assumptions based on common knowledge, my own experience, or through good advice. Wherever these assumptions are, I will point them out and explain my justifications for them. Additionally, any factual information I use will always come along with a reference to the source so that it can be verified.
One question that may come to your mind is that some of this information probably could have been easier to obtain from GW. That may be true in some cases, but in others they would probably not disclose it. I am just going to avoid that hassle and independently research everything I need. Like I said with the assumptions, some of the information will be close but not exact, and I will justify its use.
I am going to frame this article as a narrative which will follow the entire life of a particular model from its origins as a design and a lump of raw material to the store shelves. The goal will be to accurately portray how much that model costs to manufacture. After that, I will look at the bigger picture of how that model fits in the global economy and how GW’s pricing and business practices compare to its peers. Are these ambitious goals? You bet! I hope to either substantiate what people complain about or to completely debunk those people as naysayers.
The model I am choosing to follow with this narrative is the Space Marine Land Raider (see below). I am choosing this model because it is all plastic, which simplifies the manufacturing analysis; it has several sprues of components that compose the model; and it has incurred a price hike within the last few years.
There is no reason to think that GW would work any differently and there is some evidence in how some of their award-winning model designers are celebrities in their own right and are sought after by competitors when the leave the company. Even though the Land Raider model was designed some years ago, I will use current labor prices in this line of work as this article represents the current state of affairs, i.e. the proposed price hike in Summer 2006 with new models being released currently. I am also going to use salaries based on designers working in the United States as it proved difficult to find a good salary calculator for Britain.
Using the job title of Artist in the Salary Calculator provided by salary.com and using New York, NY as the location, the 50th percentile income was approximately $54,000 annually, which seems reasonable for an experienced individual working at GW in a highly skilled trade such as new product concept design. Obviously, the person probably does not spend all of their time on this one project, nor the entire year necessarily. I am going to make a reasonable assumption that the design process for the Land Raider was done by one person over the course of 4 months work. This adds a cost of $18,000 into the process.
Raw Material Costs
The mold itself is a complex engineering structure. All gamers are familiar with the sprues that the parts must be clipped off of, but it is not random as to how they are laid out. In many cases, certain parts must be orientated in a certain way as to allow a better feed of material through the mold. There is also careful consideration as to where to locate the gates of each cavity, the vent holes through which air escapes, and if overflow storage is needed to completely fill cavities as they cool.
The Land Raider consists of six sprues, all of which can be combined into a single multi-cavity mold. A good overview of general cost for the mold can be found here, along with some cost savings ideas. GW most likely outsources the production of the molds, as that can require extensive machining processes.
I will use the figures presented on the Listo website and estimate the time to build the mold at 300 hours. The mold will be complex, certainly more than 50 hours, but not so much so that it takes 500 to 1000 hours to manufacture. At $60/ hour, the mold would cost $18,000 to make. However, another aspect of the mold manufacture is the design, which would be done by a tooling engineer or a qualified CAD technician. I will cut the difference between the two salary ranges ($40,000 for tooling engineer and $30,000 for CAD Tech) and use $35,000 per year or about $17 per hour. Based on my own experience with CAD design, about 40 hours of design work is a good approximation for how long it would take. This adds another $680 into the cost.
The last piece to this portion of the equation is the actual machine, consisting of the hopper, heaters, and screw drive injection system. All of that would constitute a capital expenditure and most likely the same machines are used over and over for different projects with the molds being swapped out as needed. I imagine that as the company expands, they bring new machines online; for this model, an existing machine was used and the cost would have been long-since absorbed into the cost structure of the business, so I am not going to add any cost here. If I was looking at a whole new business unit, like possibly when they added Lord of the Rings to their offerings, then new capital equipment would have been a definite factor.
This section has used three basic assumptions. The first is that GW does their own injection molding. The second is that they outsource the manufacture of their molds. As such, I won’t be using the cost-per-unit examples listed on the Listo site, as that is for their complete service, which includes the molding process as well. The last assumption that I will make here is that the mold will have a very long life relative to the product life. That means the mold will not need to be replaced—a reasonable assumption for plastic molding, as the wear and tear is much less than metal molding.
and Other Costs
There are several additional costs that I am not going to detail here. One of these is the actual cost of the retail space. This actually only applies to GW stores, as that is cost incurred to them, but since the same price is paid at both GW and non-GW stores, I am going to assume those costs are eaten up elsewhere in their business. This is also abundantly true with the sheer number of online sales replacing in store sales. The cost of running their injection molding machine, such as power consumption, is negligible on a per-package basis, as is the cost of warehouse space, especially if they run a lean operation that minimizes warehousing costs. Lastly, I am also ignoring any kind of excise or customs fees, as GW has moved enough of its production facilities into the United States.
How Many to Make?
As more are produced, some costs, like raw materials and shipping, remain constant, while others, like initial design and equipment, actually decrease on a per-unit basis. However, this doesn’t mean that as many as possible should be created, because an overproduction will actually raise costs, namely in warehousing.
The one sticking point in writing this section is that no GW rep in his right mind would tell me how many Land Raiders they sell a year; therefore, I haven’t even bothered asking. Instead, I am going to approach this part of the analysis from the other end, with some good old fashioned reverse engineering.
Let’s start with their current sales price of $55.00. According to Games Workshop’s listing of retailers, there are 865 stores in the US that sell their product. A safe assumption would be that these stores would sell an average of 20 Land Raiders the first year they were out and possibly about 10 per year after that. Using five years as how long they have been around brings the total number sold to 51,900, with total revenue of $2,854,500.
a tabulated calculation of the total cost of producing these models over
the 5-year period and then a breakdown of the cost per model. The labor
cost was calculated from the per-hour unit cost and computed over a 40-hour
week for five years. My assumption here is that GW is not a high quantity
facility that will run multiple shifts for 24-hour, 7 day-a-week production
shifts. They may have increases around the holidays or new releases, but
they are not making automobiles or toothpaste. The costs for packaging,
shipping, and raw plastic were calculated based on a cost per model, multiplied
by the number of models total.
This result places the cost per model at $8.48, or 15.42% of the sale price. But before anyone starts rending clothes and grating teeth, this story is still not finished. During my research, I found a very good article (with the exception of no references) about the cost involved in producing an adventure roleplaying game. I feel the author did an excellent analysis of the costs of production, marketing, and sales and gives some nice insight into the exact kind of analysis I am attempting to do here. The difference is that a boxed game like that is far less costly to make than a model like this, as can be seen in the comparative costs, but note how the costs of producing each of these products as a percentage of sale price is similar.
The reason I am citing this article is it gives a good look into the other costs I have thus far not taken into account, such as the cost to the distributor and the cost to the retailer to carry the product. These two things would be impossible for me to independently research, as the amount GW pays to Alliance, their main distributor, would be confidential. Also, the fact that they both have their own stores, meaning no cost to the retailer, as well as independent stores, meaning there is a cost, complicates that calculation greatly, even if I could find out for sure what the discount to the retailer is.
So for the next part, I will use the percentages quoted in that article, but I do warn that all of his numbers are simply stated with no references other than his own experience.
Starting with that author’s stated 50-10-10 rate to the distributor, this sale price of $55.00 puts the net sale price at $22.28 after the fees paid to the distributors. I will skip production costs, since the previous sections covered that topic in much greater detail. I will use the same 7% for figuring what gets paid to the sales reps. Taking $22.28 as the wholesale price cuts the net revenue down to $20.72. For advertising costs, I am going to use the low end of the scale at 5%, because GW is a well-known brand and would not have to put as much effort into the advertising of a new Land Raider. Five percent of $22.28 is $1.11 and the net revenue is now $19.61.
I have already covered packaging, and I am not going to include interest, as GW would not be borrowing for this nor would the lost interest revenue be much of a factor in terms of their cash flow. The biggest reason for this is they have a large selection of current products that are pulling in revenue as they are launching this product. In the cited article, the author is using a brand new company with no exiting products.
Adding It All
Citing “Why Games Cost What They Do,” I could use either of the figures that Greg Costikyan suggested. The first suggestion is the overhead cost is the print cost, or production cost in this case. Subtracting the packaging and shipping costs, the overhead on a per model basis would be $4.98. The other suggestion is 30% of the expected income or wholesale price which comes to $6.68 per model. Using one of those two suggestions, the profit per model would be somewhere between $4.45 and $6.15. With the estimated volume of 51,900, the net profit would be between $230,955 and $319,185.
So there you have it! In my opinion, a net profit per model of $4.45 to $6.15 seems reasonable and appropriate, but I did also state in the beginning that I would try to determine if that much of a margin was justified, as well as the price increases.
Throughout the world of manufactured goods, there is a wide range of products available but many can be thrown into one of two categories: necessities and luxuries. Something like clothing (not Tommy Hilfiger mind you) would be a necessity, while small toy soldiers would be a luxury.
If you look at a comparison of the sales figures between necessities and luxuries, the sales volumes of necessities would be far higher. However, the profit margins would be far lower. This is simply an effect of supply and demand. Necessities have such high sales volumes because everyone needs them…well, duh. Because there is such a large potential market with high volumes, everybody wants a piece of that market, so there are a plethora of companies making things like clothes, cookware, toasters, furniture, etc. Because there are so many manufacturers in these markets, they are all competing for sales and the best way to do this is to offer lower prices through cost-cutting at every opportunity. Hence, the profits in these industries tend to be near margin, with no room for error.
The opposite is true in a luxury industry, such as waterbeds, automobiles, boats, and little toy soldiers. There is a smaller market with fewer manufacturers and less competition. Thus, you will see a higher profit margin. In fact, many companies tend to do a little of both. Think of how General Motors offers both the Chevy Cobalt and the Cadillac CTS-V. A company will offer cheaper goods with less profit margin that make up a large number of sales as well as a luxury or upgraded version that provides a higher profit margin with lower sales volume. The luxury item will tend to provide more profit, while the cheap good provides stable baseline sales revenue.
But I digress. In the manufacture of a toothbrush, I expect a profit margin of $0.05 to $0.10 per toothbrush. They only cost a few bucks and they are sold in the millions. With toy soldiers, with the much lower sales volume, I expect there to be a higher profit margin. The company itself has to be taken into account as well. With a smaller company with much less market presence, such as Steve Jackson Games, the profit margin is probably a lot lower, because they don’t have a stable base of sales to provide economic security. A larger company like GW can assume a larger profit margin because they have the market presence to absorb it. Does that mean they should? Well, that is an entirely different debate on business ethics. As Kenton (and others) have stated, playing with miniatures is a luxury and choice past time, not a necessity.
I will close out this article with one last thought regarding the recent price hikes. The major resource from which plastic is made is petroleum. From 2003 to 2005 the price of oil nearly doubled, from $28.74 to $54.81 per barrel. That is a 90% increase. Obviously there won’t be a direct correlation between price of crude and plastic, but as the price of crude goes up, there will be an impact on plastic.
During that same period, GW had one major price hike and a model like the Land Raider was increased from $50 to $55, which is a 10% increase. It could be reasonable to suggest that the 90% increase in crude oil price translated into a 10% increase in Acetate Cellulosic. It is also worth noting two additional costs involved in making this product. One is shipping; most of the shipping is done by good old-fashioned diesel-guzzling trucks. The other is power consumption in the form of electricity. Chances are that most of the electricity used to power the factory is from a coal plant, and the price of coal has also been high the past few years. This results in a higher electric bill. Additionally, the price of natural gas has been skyrocketing for two reasons. The first is that it tends to follow oil, and the second is that it is a clean alternative for power generation.
has been presented here gives a good background to the next debate over
the looming Games Workshop price hike.
© copyright Robert Van Pelt, April 2006. Used with permission
Codex <> Tactics <> Gallery <> Allies and Enemies <> Tales of the Tigers