Assessing Three Statistics in Support of Dodd-Frank 1502

In the debates over conflict minerals and Dodd-Frank 1502, a couple statistics are frequently used to point to Dodd-Frank’s success.  I did some research into these statistics, tracing back citations and looking through reports to assess the significance of these statistics and what they say about the success or failure of Dodd-Frank 1502.  I decided to compile my findings into this blog post, in the hope that it can be a public resource that supports informed debate about conflict minerals.

The three statistics I examine are perhaps the most commonly used statistics used in campaigning supportive of Dodd-Frank 1502.  They are:

  • “Armed groups and the Congolese army are no longer present at two-thirds (67 percent) of tin, tantalum, and tungsten mines surveyed in eastern Congo’s North Kivu, South Kivu, and Maniema provinces.”
  • “As of May 2014, nearly three-quarters (74 percent) of 3T miners were working in mines where no armed group involvement has been reported.”
  • “The Dodd-Frank law and electronics industry audits have created a two-tier market for tin, tantalum, and tungsten (3Ts) from Congo and the region. Minerals that do not go through conflict-free programs now sell for 30 to 60 percent less, thus reducing profits for armed groups trying to sell them.”

To examine these statistics I primarily used three reports: the one in which they originated, the Enough Project’s June 2014 report, “The Impact of Dodd-Frank and Conflict Mineral Reforms on Eastern Congo’s Conflict”; the IPIS report that provided much of the information to the Enough report; and the 2015 UN Group of Experts (GoE) report.  The Enough report relied heavily on the 2014 UN Group of Experts report, and although Enough could not have used the 2015 report as it was released well after the Enough report’s publication, I chose to use the 2015 version due to its more up-to-date information and its extensive focus on armed group revenue.  I supplement these three reports with past research on the subject.  I start by looking at the shift from 3T (tin, tungsten, and tantalum) mining to gold, then assess the three statistics, and finish by discussing a few important points about the conflict mineral trade and what a more complete perspective on the three statistics implies about the success of Dodd-Frank 1502.  Much of the information I rely on is several years old and could now be less applicable, and it is also constrained by the extreme difficulty of conducting research in the eastern DRC, though it should be noted those problems apply to the three statistics as well.

Christoph Vogel has discussed these statistics before, arguing that they are incomplete and not an accurate reflection of the situation in the eastern DRC.  This is primarily true, and much of this blog post discusses why these statistics do little to suggest Dodd-Frank 1502 is working.  Above all, the use of these statistics to support the success of Dodd-Frank is limited by the decline of 3T mining.  All three of the statistics focus on 3T mining, which now constitutes a small portion of natural resource exploitation in the eastern DRC, and as a result the statistics are relatively insignificant.  While the shift from 3T mining to gold mining can largely be attributed to Dodd-Frank, the last section of this blog post will discuss why that does not mean Dodd-Frank has been successful.

The shift from 3T mining to gold mining:

While neither Dodd-Frank 1502’s supporters nor its opponents have devoted a great deal attention to this shift, a major consequence of the legislation and related efforts to reduce the purchase of conflict minerals has been the decline of the importance of 3T mining and the increase in the importance of gold mining.  IPIS writes that “while 3T mining used to attract more miners…gold mining is currently by far the most important subsector in Eastern DRC’s artisanal mining sector… Out of an estimated 221,500 artisanal miners active in the mining sites visited in the framework of this research (1088 in total)… four out of five artisanal miners work in the gold sector.”  The GoE report tells a similar story, estimating that annual net profits of gold are $40-120 million while just $7.5-22.6 million for 3T minerals.  IPIS points to multiple reasons for this shift: “international regulation, low demand, low prices paid for the minerals, the growth of a semiindustrial mining sector and the depletion of certain deposits.”  Dodd-Frank 1502 and other traceability efforts are certainly major factors, and the growth of gold mining makes sense in this context.  While conflict-free regulations have put a major dent into the 3T trade, gold’s small size makes it easier to circumvent regulations.  It is hard to overstate just how much gold evades regulations; the GoE report estimates that official gold exports constitute just 2% of the actual figure, while 98% is smuggled.

As a result, the use of only statistics on the 3T trade reveals relatively little about the mineral trade in the eastern DRC and the success of Dodd-Frank 1502.  Regulations have stifled the 3T trade but have struggled to contain the gold trade, and it should be noted that Enough has acknowledged these problems.  Seeing the difficulties of the 3T trade, both miners and armed groups now focus more on gold as the trade seems almost entirely unaffected by regulations.  With the minor importance of the 3Ts in mind, I will now address the three statistics.

“Armed groups and the Congolese army are no longer present at two-thirds (67 percent) of tin, tantalum, and tungsten mines surveyed in eastern Congo’s North Kivu, South Kivu, and Maniema provinces.”

There are several shortcomings with this statistic.  First is the use of the phrase “no longer present,” implying that at one point armed groups were present in all the mines.  Enough does argue this, almost always referring to the same sentence originating the 2010 GoE report: “in the Kivu provinces, almost every mining deposit is controlled by a military group.”  The report does not include a great deal to support this claim, nor is it a major focus of the report, but says the finding is based on MONUC mapping exercises.  One major source of armed group involvement reported is the CNDP, which disbanded before the implementation of Dodd-Frank and was succeeded by M23, which did not seek to control mines.  Also, it should be noted that the report finds the omnipresence of armed groups at mines in the Kivus, while saying that armed groups are present to “a lesser extent” in Maniema.  The 67 percent statistic now used by Enough includes Maniema as well as the Kivus, and even the GoE report the Enough Project cites denies that every mine in Maniema was once controlled by an armed group.  As with all these statistics, the correlation between Dodd-Frank and reductions in armed group presence does not mean causation, as other factors, such as the improvements in the Congolese military and the use of FIBs, could be behind the changes.

The statistic also relies on the IPIS survey that, while very thorough, is incomplete.  It surveyed 1,088 mines.  There are at least 800 mines in South Kivu alone, and while the figure is not known for North Kivu, between the Kivus and Maniema there were certainly many mines that were not surveyed.  It also seems logical that mines controlled by armed groups would be less likely to be surveyed, as increased security concerns would make them more difficult to reach.  In addition, as with all statistics on armed group presence, it is not exactly clear how reliable the statistic is.  The IPIS report does not detail exactly what “present” means.  Though it includes some forms of taxation, it almost certainly does not include taxation at roadblocks, which is one way armed groups have sought to get around increased regulations.  Given the direct and indirect ways armed groups benefit from minerals in a militarized economy and the fluid movement of armed groups, it is very likely armed groups could have benefited from a mine in a manner not visible to IPIS surveyors, especially as their visits to each of the 1,088 mines could not have been particularly lengthy.  Finally, as with the other statistics, the 3Ts are not all that significant.  Dodd-Frank 1502 may have led to low armed group involvement in 3T mines because of the difficulty of acquiring conflict-free status, but looking beyond just 3T mines, IPIS found armed group involvement in 79% of all mines in North Kivu, 58% in South Kivu, and a small amount of mines in Maniema.  The 2015 GoE report estimates an armed group presence at 57% of mines in the eastern DRC, while around 25% of mines when just looking at the 3Ts.

“As of May 2014, nearly three-quarters (74 percent) of 3T miners were working in mines where no armed group involvement has been reported.”

This statistic is probably the most meaningless of the three.  As with the above statistic, it is limited as IPIS did not visit many mines and it would have been very easy for armed group involvement to evade the IPIS researchers.  Similarly, the statement that “no armed group involvement has been reported” does not definitively mean there was no armed group involvement, just that it was not reported to IPIS researchers.  The statistic is relevant from a worker safety standpoint and while that is important- conditions in Congolese mines are truly horrific-, it reveals little about the impact of Dodd-Frank on ending the conflict.

Yet the statistic also reveals relatively little about conditions for miners.  Perhaps the best explanation of why this is the case comes from the discussion that immediately follows the statistic in the IPIS report in which it originates, though this discussion has gained far less attention than its first sentence:

“The data on our map even shows that almost 74% of 3T miners are currently working in mines where no armed group involvement has been reported. As discussed above, however, the number of 3T miners has declined, as many of them have turned to gold mining. Consequently, less than 20% of artisanal miners are working in the Eastern Congolese 3T sector. The positive developments in the 3T sector stand in contrast to the gold sector. Gold mining in Eastern DRC is plagued by security problems and fraud. According to the data gathered during our teams’ field visits, 57% of gold miners are working in mines with armed group presence.”

“The Dodd-Frank law and electronics industry audits have created a two-tier market for tin, tantalum, and tungsten (3Ts) from Congo and the region. Minerals that do not go through conflict-free programs now sell for 30 to 60 percent less, thus reducing profits for armed groups trying to sell them.”

While this statistic seems to be valid, it is not clear this is good news.  First, as stated multiple times, 3Ts constitute a small portion of the mineral trade, and while there may be a two-tiered market for the 3Ts, there is not for gold.  Yet additionally, although certified conflict-free minerals may sell for more than uncertified minerals, they both seem to sell for less than the original price.  The 2014 Enough report cites that in 2009, tantalum, conflict-free or not, sold for $132/kg, but in a later section of the report it cites, seemingly positively, that now conflict-free tantalum sells for $40-45/kg and untraceable tantalum sells for $25-35/kg.  The global price for tantalum actually seems be higher now than in 2009, with a price of $122/kg.  While I have not seen reports of such dramatic drops elsewhere, the figures in the Enough report are worrying.  The Washington Post reports that tin previously sold for $7/kg and now non-certified tin sells for $4/kg, while Enough finds that conflict-free tin sells for $5-7/kg.  The global tin price has risen since the implementation of Dodd-Frank 1502.  It seems Dodd-Frank 1502 has led to lower competition and for companies to attempt to compensate for the price of tracking their supply chains, resulting in lower mineral prices, conflict-free or not.

And while the statistic focuses on the fall in armed group revenue, a significant effect has been the fall in the revenues of miners.  Whether or not the conflict-free price is also lower than the original price, a two-tiered market does not suggest clear benefits to the majority of miners.  Even though most miners do not contribute to armed groups or do so because they have no other choice, these miners still struggle to benefit from the higher prices for conflict-free minerals.  Only 141 mines in the eastern DRC have been certified as conflict-free out of well over a thousand, meaning that the vast majority of miners suffer from the lower prices for 3T minerals just because of the slow pace of the certification process, a key consequence of the “guilty until proven innocent” approach of Dodd-Frank 1502.  As a result of the inadequacy of conflict-free certification processes to cover any significant portion of the Congolese market, it should be noted that the argument for Dodd-Frank 1502 is often implicitly that a “de facto embargo” is better than the alternatives.  With conflict-free products rare and even then not necessarily actually conflict-free, the idea that the Congolese market can be accessed and mineral profits can be largely stopped from going to armed groups is barely foreseeable in the near future.

Since the number of 3T mines controlled by armed groups has fallen while armed group control is much more common in gold mines, the Enough Project’s arguments would suggest more miners should move into 3T mining.  Instead, following the implementation of Dodd-Frank, miners have overwhelmingly moved away from 3T mines towards gold mines.  Despite the greater armed group presence in gold mines, it seems miners are driven away from 3T mines due to the extreme difficulty of achieving the conflict-free label and to the gold trade where smuggling is much easier.

Although the Enough Project finds that former 3T miners have been able to “find other work and generate income,” this is not particularly convincing.  The three primary sectors Enough points to are “gold mining, farming, and small business such as fishing or motorcycle taxi driving.”  As discussed, gold mining suffers from extensive armed group involvement.  Enough says that motorcycle drivers can make as much as five times as much as they did mining.  However, it seems few miners would have enough start-up capital to buy a motorcycle and business would suffer in a service trade in the eastern DRC, especially when the largest industry, mining, has been disrupted.   And as with farming and fishing, if the profits and conditions are much better, one wonders why the miners were not in those sectors in the first place.  Finally, due to the militarization of the economy and the variety of taxation schemes, ex-miners could very well still contribute financially to armed groups even when in other sectors.

Are Conflict Mineral Regulations Working?

The three statistics alone certainly do not provide evidence that Dodd-Frank 1502 and related traceability efforts are working.  However, while we know the 3T trade has fallen, a key question is how much other sources of revenue have replaced it.  The implied argument of proponents of conflict minerals efforts seems to be that this revenue has not been replaced, though they have rarely fleshed it out.

This is an exceptionally difficult question to answer, not least because of the wildly varying figures one can see between credible estimates on the volumes of resource trades, and the answer to this question would best be answered by someone with more expertise than I have.  For me, it seems highly unlikely the gold trade itself has made up for the 3Ts.  The 2015 GoE report’s estimates of gold production are fairly similar to the ranges of estimates prior to 2010, rather than rising substantially to compensate for the fall in the 3Ts.

Additionally, IPIS found 221,500 miners in their survey.  This is almost certainly an underestimate, as they did not survey all mines and it seems likely they recorded fewer miners at the mines they visited than the actual figure.  However, this figure is well below past civil society estimates of the number of miners, suggesting that the gold trade is a long way from compensating for the fall in the 3T trade despite its gain in relative importance.  This is in line with the Enough Project’s finding that some miners have moved from the 3Ts to gold mining while others have moved away from mining altogether.  However, it is also in line with arguments of a de facto embargo on Congolese minerals and huge unemployment among miners.  Congolese minerals are now certainly very difficult to sell due to the difficulty of obtaining conflict-free status and low prices, and while gold can circumvent these problems due to smuggling, the 3T trade has certainly taken a major hit.

I do not believe these findings suggest traceability efforts are working.  I think it is nearly unquestionable that armed group revenue is lower now than it was before Dodd-Frank 1502.  However, especially as the 2015 GoE report estimates that organized crime makes more money from other sources of revenue- such as charcoal, timber, and wildlife- than the 3Ts and gold combined, and there does not appear to have been a dent in the gold trade, it does not seem armed group revenue is dramatically lower than it was before Dodd-Frank.

Proponents of Dodd-Frank argue that it is has succeeded in diminishing the 3T trade and now additional efforts must be made to combat the gold trade.  Yet it is not clear that the gold trade can be combatted.  The ease of transporting gold due to its small size and the complete failures of any efforts to eliminate smuggling cast doubt on the possibility of regulating the trade.  Even with the 3Ts, there are serious doubts about the accuracy of minerals labelled conflict-free, especially with the problems in the bag and tag system.  Additionally, it seems highly unlikely that charcoal, timber, and even general taxation of the local population- as M23 used extensively- could be eliminated.  We also have not seen evidence of armed groups disbanding or being particularly weakened as a result of Dodd-Frank 1502.  It should be noted that minerals were not a primary cause of the conflict, and local conflicts over land, conflicts of identity and citizenship, and an absent state are major causes that will not be resolved by reducing mineral revenues.

The group that we do know is suffering from losses in revenue is miners.  The reduction of armed group presence in 3T mines has done little to help miners.  Those remaining in the 3T trade suffer from low prices and exclusion from the market.  Many have moved on to the gold trade where armed group presence is high, many others have been forced to move into other sectors that one would guess are less lucrative, and it seems likely that hundreds of thousands are unemployed.  In a context with almost no safety net and where miners have many dependents, I seriously doubt that the seemingly moderate reductions in armed group revenue lessen human suffering more than the costs to miners increase it.