Proliferation for profit is in the air. You can hardly turn on your TV or radio without seeing David Albright talking about his excellent new book, “Peddling Peril.” I have long advocated thinking about proliferation as a business and applying the results of research on the diffusion of technology to understanding how countries get the infrastructure and know-how for making WMD and the means to deliver them. As such, it would be nice to understand the amount of profit a proliferation profiteer like A. Q. Khan or North Korea makes on selling various technologies the world would rather see end. If we could understand the amount of profit to various forms of proliferation, say selling missiles vs. missile factories, we might be able to understand how the nature of proliferation is changing.
Unfortunately, it is very hard know the profits associated with any given sale or even industry. It is even hard to get firm numbers of the gross transaction much less the profit made. I have tried to assemble a collection of prices for various missile component, system, and production plants that can be used to give a feeling for the order of magnitude of the problem. The ones I feel relatively confident in are shown in the table below. Prices have been adjusted to be in 2010 dollars.
| Missile Component/System/Plant | Price (in 2010 Dollars) |
|---|---|
| 1 Iraqi Badr-2000 missile | $9,900,000 |
| 1 SCUD-B plus TEL | $220,000 |
| SA-2 Sustainer Engine | $14,000 |
| Badr-2000 Factory | $185,000,000 |
The Badr-2000 was a virtual missile sold to Iraq in 1984 by a consortium of countries that were hoping to use Iraqi financing to develop a missile that was being sold by as “equivalent to the Pershing II.” Since the Badr-2000 was never actually built, it is possible that these numbers are unrealistic, though you could make an argument either way as to whether the actual price would be more or less expensive. It does establish a ball park number for a sophisticated, long-range missile production facility. The number I am using for the SCUD-B comes from reports on what a “collector” in the US is reported to have paid for a fully militarized missile and it’s TEL. It is hard to imagine this collector paying less than the price of such a deal would be on the “open” market. Nevertheless, I would love to have other examples of SCUDs being sold to check this. (Hopefully without the TEL. Of course, any well documented numbers for missile prices in the developing world would also be greatly appreciated.)
Let’s try using this table on a media report from 1992. Bill Gertz reported on March 10th of that year, that North Korea shipped a complete ‘SCUD’ factory to Syria for a reported $300 million. (Again, this is in today’s dollars.) Of course, this was about the time that North Korea was shipping Nodong factories to Syria, Iran, and Pakistan so it seems quite possible that this was actually a Nodong factory. The “misidentification” is, of course, quite reasonable since Nodongs hadn’t become as well known as they are now. Also, they are, after all, derived from SCUD technology. But the point I want to talk about is: how believable is this price? Comparing it to the Badr-2000 factory complex, it seems too high. After all, the Badr-2000 was supposed to be a very sophisticated missile with both solid- and liquid-fueled stages.
I am, however, inclined to believe it as a ballpark figure though not enough to include it in my table. If true, it would seem that today’s market is willing to pay a premium for the know-how to build a larger, more powerful missile. If we consider the costs to North Korea in developing the production line to be sunk costs, then this know-how is pure profit. This might explain why North Korea is willing to sell factories to proliferators as opposed to insisting on selling missiles. Of course, there are a number of other reasons for selling factories (with a “small” number of example missiles). For instance, it is possible that North Korea is doing exactly what Iraq’s Badr-2000 suppliers tried to do: get other countries to pay for the missile’s development.
Note Added:
“Know-how” is a technical trade term that refers to a written description of the product and process for its manufacture. Wikipedia has an article on it though they seem to also include the tacit knowledge needed to operate the machines. I disagree with that somewhat since all the contracts that I have seen have made a distinction between know-how and training.
I disagree with your assertion that the 300 million for a missile factory is “pure profit.” I presume that they would have to deliver the tooling for that and producing this tooling would require the diversion of industrial resources that might have been used for other things. Unless, of course, they are decommissioning an existing plant and selling it secondhand.
It seems more likely to me that North Korea prefers to deliver whole plants because the tooling is plausibly dual use and less likely to be intercepted en route.
I also think it’s very possible that the collector paid less than the open market rate for a SCUD. It may have been stolen, in which case the collector got a five-fingered discount.
Steven Zaloga puts the export price of a Scud-B missile in the late 1980s at one million dollars (Scud Ballistic Missiles and Launch Systems 1955-2005 p.39)
Regarding the Syrian factory, I don’t think it was a Nodong factory. Due to its proximity to tiny Israel Scuds are pretty much sufficient for them. They have acquired a number of Nodongs later on, so they can reach Irael even from within the most Northern parts of their country but in general the Nodong is a bit of an overkill.
James,
I said transferring “know-how” was pure profit, though perhaps I didn’t fully explain what “know-how” is. Let me take this opportunity to clarify it. The phrase is a standardized trade term that means a written description of the product and the process for manufacturing it. Today, that could all be contained on a 25 cent CD so ignoring the cost of FedExing it (and the 25 cents) its pure profit. The cost is less than $300 million but that is a subject for another post.
I would need to see more evidence than Mr. Zaloga’s book before I believe $1 million for a SCUD. As for $300 million for a SCUD factory. That was one of the reasons I quote the cost of a SA-2 Volga engine: $14,000 (a number I am certain of). That engine and a SCUD simply dont differ all that much so a price of $1 million dollars just seems far too high.
I’m sorry; I was conflating the two paragraphs. You’re right, of course, in that intellectual property is nearly all profit. But on a 300 million dollar factory deal, I think the know-how is really just a tease to sell the equipment. As you say, it’s all sunk costs, anyway.
North Korea’s decision to sell entire factories may be driven by a cold analysis of the missile market. Proliferating entire factories may make offer less profit than selling individual missiles, but that’s over the long run. Over the short run you make a large profit up front and the problem of the SCUD’s obsolescence is transferred to your customer.
At the time of the purported Syria deal China was marketing their M-9 and M-11 missiles and Russia’s economic problems seemed poised to release unlimited numbers of aerospace engineers-for-hire onto the global market. The North Koreans may have decided the future value of their SCUD line was declining in the face of Chinese competition. Offering to sell the production capability would seem to be a marketing response to the introduction of a superior product by a rival.
yes, the nuclear industry is very capital intensive. so there will always be a premium to innovate in order to lower capital costs.
But lowering capital costs is bad from a non-proliferation perspective.
Uranium enrichment seems to be peoples exhibit A for the effect. Gas centrifugation and now laser enrichment (i.e. silex) have lowered capital costs but in a way that poses challenges for the non-proliferation regime.
I’m doing research on this myself, but from the perspective of the economics of externalities.
I kinda see proliferation as an externality. That’s why we have regulations like safeguards for in a perfectly free market in nuclear technology it would not be rational to factor in proliferation concerns; it’s a cost.
Geoff:
A meaty choice of topic.
I’ve been doing some research lately into the market for North Korea’s ballistic missile exports – stay tuned! – and wound up not using any of the dollar figures that I came across. The sourcing and accuracy are hazy, and numbers only crop up so often. It’s not so easy to fashion a coherent picture out of these dribs and drabs, so I’ve set this material aside.
So here, for whatever they may be worth, are the figures that I’ve spotted in assorted reports and documents for various offers, transactions, or expenditures, but not used. Some relate to prices (e.g., what did North Korea or AQ Khan charge for this or that), and some relate to costs (e.g., what did it cost the North Koreans to launch a rocket). I’m sure there’s more out there, yet to be unearthed.
Caveat lector: Rather than sweat the price deflators, I’ve simply provided the date of publication. (Note that this is not always close to the date of transaction or expenditure!) Shifts in exchange rates are also disregarded, when figures are given in non-USD currencies.
1. An unknown number of Scud-C missiles from North Korea for Peru.
2. Ten Nodong missiles and a Nodong factory for Iran.
3. Ten Nodong missiles and a Nodong factory for Pakistan.
And:
4. Downpayment on a Nodong factory for Iraq.
5. North Korean missile sales – numbers and types not stated.
6. A shipment of conventional weapons from North Korea to Iran:
7. North Korean weapons sales over long periods.
And:
And:
8. Scud-B missiles from Libya.
9. Estimated costs of North Korea’s 2009 missile and nuclear tests.
And:
10. Costs associated with Pakistan’s nuclear weapons program.
And:
And:
And:
11. Price of the AQ Khan network’s services.
And:
And:
And:
12. Compensation received by members of the A.Q. Khan network:
Geoff, Josh,
Is a Scud sufficiently comparable (technologically) to some older missile from France, UK, USA? If so then maybe it’s possible to use actual prices from these countries’ historic defence budgets to put some bounds on what it costs to produce a Scud. Admittedly production technologies have moved on since a few decades ago, but I’m guessing that the answer may still be within a factor of five or so. I suppose if one knew how much skilled labour it takes to build a Scud then – on the assumption that material costs remain about constant relative to the rest of the economy – it might even be possible to refine the estimate using a labour-cost price escalation.
Missiles are out of my expertise, but it would certainly be handy if there were some way to reject patently implausible reported values in Josh’s summary, and then maybe just average the rest.
Hairs,
Unfortunately, being within a factor of five isn’t very useful. It wouldn’t, for instance, see any difference between the $220 K for a SCUD that I derive from reports about what a California collector paid for a SCUD (and TEL) and the $1 million quoted in the Zaloga book. However, all the other numbers in my table have a high level of confidence.
The problem with some of the numbers Josh has quoted is that it is uncertain exactly what is being referred to. Take the Khan $50 million. He says it was for “technology.” That could mean for know-how. If so, it is probably fairly reasonable. That implies that Pakistan thought it was sufficiently familiar with production processes etc that it could design its own production line. Pakistan (and India) did almost exactly that when they licensed solid propellant technology from France for producing sounding rockets. In that case, however, France did supply a list of production equipment to be used as well as a list of suppliers for that equipment. That too, might be possible for North Korea to do and to include in “know-how.” (I think this is what proliferation will look like in the future, but again, that is the subject for a future post.) Then again, Khan could have meant production equipment when he said “technology.” In that case, it is unreasonably low. It is that ambiguity in many of these reports that makes them hard to use.
I’d be careful to distinguish the manufacturer’s costs from the buyer’s price. As we see in the Khan network examples, as in economic life more broadly, there can be considerable mark-up. But we can at least see what’s in the ballpark.
Somewhere out there, there are bound to be price lists issued by the North Koreans, just as there were quotes issued by Khan. But I wouldn’t know where to look to find those.
Re: Khan, I omitted one of the more interesting details from the “Project A.B.” document that appears on the ISIS website: $5 million for a bomb design!
Here’s another item of interest:
Josh,
Thanks very much for your thoughtful and very informative comments. I look forward to the publication of your study on North Korea’s role in missile proliferation around the world. It will be an important contribution to the field on proliferation studies.
North Korea has sold at least 490 Soviet-style Scud missiles to Pakistan, India and Middle Eastern countries since 1991, a South Korean scholar said Wednesday.
India did not buy a single missile from North Korea. It already had the indigenously-built Prithvi missile.
Politically speaking, missile proliferation provides the leverage for North Korea against the US. This kind of leverage has usually been in the hands of USA, Russia & China. It forces the USA to play nice with North Korea.
It’s almost impossible to figure out what “the cost” is to manufacturers of the missiles – labor rates vary so much, absent active commercial efforts to amortize capital and industrial investment the overhead to be applied per unit is often unknown, etc.
The experience from the Russians entering the space launch market pretty much made it clear that the costing / pricing for non-fully-commercial parties is essentially monopoly money. Someone’s getting paid, but you need to see the full books and understand all the system effects with the way that nation’s economy is managed to see what’s really going on.
Well, they haven’t necessarily been playing nice, but the point is taken: not all proliferation is for money. If you want to keep the firemen busy, you set lots of fires. Sharing the weapons creates a global constituency that opposes non-proliferation efforts, which can be priceless in itself. In such cases the money exchanging hands is a mere formality based on the presumed ability to pay, the desire of the seller to attach the buyer politically or economically, and the marginal strategic value of the weapons themselves, rather than their marginal production cost.
Analyzing the sale of major weapons as a purely free market may not be entirely useful. In far more open exchanges of advanced weapons there are frequent accusations of unfair competition, politicization, and bribery. Have we become so cynical that we now believe the North Koreans and Syrians are more logical and professional in their arms deals than we are?
The missiles are almost never delivered on a standalone basis. They are usually part of a defense package (like US) or a barter for another technology (North Korea & Pakistan).
The money exchange is the difference between the value of the exchanged commodities.
Rubin,
Even if that were true, which it is not, they would still have a price both sides agreed to. Otherwise, they would always be trying to solve questions like “How many oranges for a SCUD?” Money is a very useful concept, even when it is totally abstract.