Side event – Understanding drug markets: An economic perspective

Side event organized by the Government of Switzerland

Switzerland: Welcome. It is key that we exchange between stakeholders. The aim of these meeting is to offer a platform to be able to work together in the coming few months, leading up to the High Level Ministerial Segment. Some of the panelists today might not align with our national position, but we think their work is important in our quest to understand demand and supply. We start with presentations on supply and then move on to talk about demand. Let me introduce Peter Reuter, professor in the School of Public Policy and in the Department of Criminology at the University of Maryland. He is practically a cofounder of the discipline and a co-author of six books on drug policy. He was the first president of the International Society for the Study of Drug Policy. He worked on assessing Swiss drug policies in the 90’s.

Peter Reuter, Professor in the School of Public Policy and in the Department of Criminology at the University of Maryland: I appreciate the chance to present today. I’ve been publishing for 25 years and one of the major observation I made is that policy is not framed in an economic way when we talk about demand and supply reduction even though they are the mainstays of discussion. This reflects the fact that drugs are not exclusively distributed in classic market contexts, they are exchanged in barter, social networks and informal relationships. The two drugs that have been the objects of my attention are heroin and cocaine.

If heroin costs 1 dollar/gram, the consequence would be very different from the situation now. So it’s important to understand what makes the prices. The supply curve is basic economics and the prevailing notion is that enforcement shifts the curve to achieve higher prices and through that, lower consumption. That is a nice clean story you can find in basic economics texts and a useful example if you want to entertain college freshmen. Supply-side enforcement shifts supply curve, raising equilibrium prices and reducing equilibrium quantity. Impact on price and quantity depends on: relative price-elasticity of demand-side of the market, effectiveness of enforcement measures on target activity, group of people, or resource. Substitutability of that vulnerable activity, group, or resource in supply activities in this market. Since the supply side is relatively elastic, consumers would bear most of the resulting burdens. Enforcement raises prices, and raises profits for some suppliers. The starting point for me today is that supply reduction is not a single programme. Drug supply reduction efforts range from crop eradication efforts to restricting access to precursor chemicals, interdiction against smugglers, arresting and imprisoning high-level dealers or arresting and imprisoning low-level drug-sellers.

So how does enforcement affect supply? It makes risks and costs greater. How do we work out effective enforcement then, and whose risks and costs are raised? Eradication targets growers. Smugglers’ interdiction resulted in a spike in prices but the result is an increase in their margins as well. These are all short-term effects. This is similar to what we know about the effectiveness of different kinds of programmes. Let me give you a flavor of those effects. We’ve been involved in simulation studies to produce graphs but there should be a disclaimer that this is the creative output of researchers. Empirical studies look at the affect in the area of eradication: it does reduce production but not in the long term. Production moves, so overall production is not likely to be reduced by targeting specific areas. Regarding precursor control. We have only one research group, who mainly observe the US for methamphetamine and their findings are typically that prices are effected in the short run but within a year, usually within 6 months there are adaptations made.

In 2006, a chemical version of cocaine was introduced to the market and the price increase was as high as 50%. It not only sustained, it showed up in the consumption side as well – a major decline. However, it didn’t show up immediately. So we thought it had to do something with crackdowns and the issues with smuggling from Mexico, but we didn’t see these effect on heroin. We looked at enforcement on different levels of the market and we haven’t found consistent studies. I am skeptical that these programs have an effect but to be honest, the quality of researches is not great. It is difficult to come up with a design and gather data to be cleverly figuring out identifying relations. I am also skeptical about the effectiveness of conventional enforcement because if you look at what happened in the USA, as shown on my graphs about the price of cocaine and heroin, when enforcement, judged by the number of people in jail and punished, was at an all-time high, it didn’t show a strong effect on prices. The inconsistent results on various levels of distribution can be explained by varying specification of enforcement intensity, alternative model specifications and/or the true differences across drugs.

The implication of all of this has nothing to do with the question, whether we should legalize drugs or not, it is about effectively organizing enforcement is a prohibition regime. Prohibition itself raises prices and restricts access, which is a good accomplishment, but supply side successes merit further careful analysis to understand the complex ”mechanism”. 2010 is the last published figure we have on our hands, but there is a study coming out next month.

View PowerPoint presentation.

Jonathan Caulkins, Professor of Operations Research and Public Policy at Heinz College at Carniege Mellon University: The issues surrounding Marijuana regulation has been a focus of mine and what we need to know about markets in radical times of transformation. Today I am going to talk about why prices matter. Since legal regimes don’t effect prices as much, I will focus on the things that do change that. I am an engineer so I think more in terms of businesses – what drives prices are production costs. The most exciting changes in drug issues are related to fentanyl and cannabis.

We can see that our data shows that when price goes down, emergency room cases go up. How causal is his relationship? How explanatory are prices? Before I show people this chart, people usually think addicts don’t care about prices, they have to have their fix. But a simple model in economics, the constant elasticity demand curve, will help us understand price raised to a constant elasticity can explain most variation in ED. Price is able by itself to explain the dependent variable.

Generally, we tend have results that there is no connection between cannabis price and prevalence but nobody is adjusting changes in price and potency. What about third variables? Let me use the Australian heroin draught as an example: overdose cases plummeted and if you just look at prices “raw”, you will see nothing. Australia is the world’s greatest place for purity analysis however… so we could adjust to potency, and there was indeed a change in purity, most of the change can be explained by prices, even rapid changes without the measure of poverty and societal factors. That is my quick argument: prices matter.

So what drives prices? In short: production costs. The big change in cannabis is policy liberalization. It’s not an on-off switch though, we are looking at changes over a generation. My comments retain to radical changes: inviting business-oriented organizations to be involved in production – there are other measures that don’t effect the drive (decriminalization or legalizing personal possession). What I’m talking about is industrial production. The retail prices dropped by 2% per month in Washington… What happened to potency over this time: we basically have good and crummy stuff on the market – and, in terms of flowers, the average went from the crummy stuff to the good stuff. In terms of concentrates: they are extremely pure as a standard. So… prices down, potency up. I’ve dedicated the last 6 months of my life to understand the wholesale price’s movement. In summary, it goes down a lot. Illegality forced methods of production into dodgy places, and say, would you buy cucumbers grown in a basement under artificial lights? Once you have legalization, you can grow this phenomenally productive plant freely.

Could the price fall even more? With any other high-labor vegetable, the production costs are 5-20 thousand dollars per acre, so 10 000 per acre (…10 dollars per pound) is a very reasonable yield. So it could go down to pennies per joint. Production costs can go radically lower after legalization. Within prohibition, changing the times of imprisonment don’t affect the quantities on the market much, but allowing businesses to produce can radically change that. Does this matter? Yes. Policies affect drug use but it’s not prevalence that matters, it’s about intense use – don’t target prevalence, target intensity. The people who admitted trying cannabis within the last year didn’t change significantly, but last month prevalence much more. If you ask number of days of us, that quadrupled: and daily users went up even more.

In the remaining time, I want to talk about fentanyl more, because fentanyl kills. It is crazy cheap and crazy potent. At a wholesale Fentanyl costs about 1/16th as much as heroin and is about 16 times more potent than morphine so fentanyl is 16*16 =256 times as cost effective. Shift from prescription opioids to heroin to fentanyl resulted in overdose deaths soaring. These are worrying trends!

In summary, price matters, it follows production cost in the long run, two big contemporary drug issues can be seen as altering production costs: (Legal) Cannabis and Fentanyl. Studying suppliers’ is useful – much of the literature thinks about the users, but it is also important think about suppliers too.

View PowerPoint presentation.

Now on the demand side, Alison Ritter, drug policy scholar and Driector of the Drug Policy Modelling Program (DPMP) at the National Drug and Alcohol Research Centre (NDARC) at the University of New South Wales: I will be focusing on treatment.

Let’s start with the comparison of drug budgets, we have data from 2013 (see slides) Supply reduction might not be the best investment as Peter demonstrated. Are we spending enough on treatment? I am not going to summarize all the studies on the effectiveness of treatment, but we know that it is indeed effective. The 2018 World Drug Report shows the gaps in treatment so we are missing most of the population who need our support. This notion of ‘need’ is an interesting concept. There are measurements to deliberate if an individual needs treatment. If governments wanted to close this gap and doubled your current investment, still 30% would be treated only. These are quite confronting numbers. My work was aimed at understanding the treatment gap better. This notion of need for treatment is not the best measure. Not everyone who meets diagnostic criteria wants treatment. Some people go to self-help organizations, find love, careers, families, find religion – they treat themselves, but there are still a large population that needs systematic support.

Let’s discuss demand for treatment. People who qualify as in need and are willing to seek treatment. This fluctuates. It is not just the intention of the people in need, the quality of available services also matter. If you have a bad system, or offer one kind of treatment, you will not have good measurements of use. If you look at the literature in treatment demand, very low % of people met demand.  Actually we are doing way better in meeting demand when it comes to illicit drugs – alcohol tobacco are really harmful and go without much response. The usual treatment types are interventions and screening, detoxification, counselling, medication (incl OST), residential rehabilitation and therapeutic communities. Actually, our notion should be updated and an intersection with social welfare services should be considered. We have to talk about: How is treatment defined? What is the role of primary health care services, self-help, and online, new technologies? Where do the intersection with social-welfare systems lie? What are the locally culturally specific practices?

Four key features in current research:

  1. Where does alcohol fit in? Should alcohol be separated out, or combined with drugs? Most countries have a dedicated government unit for AOD treatment policy (66%) A small # of country’s have a separate drug treatment government unit (7%) Most countries offer predominantly combined/integrated care – EG: Australia all combined care, including OST
  2. Disease management vs episodic. M ost nations currently have ‘episodic’ model. Researchers and practitioners calling for disease management framework (chronic, life course approach). It’s difficult to shift systems that are reliant on “episodes of care”. In Australia, funding is based on each ‘episode of care’. So one of the messages from me today that how we chose to fund treatment, actually then represents how we view persons in need of treatment. The attitude of the nation can be seen in their purchasing arrangements re treatment.
  3. Purchasing arrangements – How treatment is funded & purchased drives significant reatment policies: EG Health vs social-welfare, Disease management or episodic. Treatment settings: hospital, community, NGO. “Competitive tendering” associated with social-welfare or “Block funding” associated with community long-term services or “Unit cost funding” (episodic) associated with hospitals
  4. Integrated with health or separate systems? It’s a huge debate. It depends on view of the “problem” (as disease, delinquency, or social problem) If ‘delinquency’ = control (criminal justice systems) If ‘disease’ = health care, If ‘social problem’ = social care. A nation’s shift (depending on their history of drug policy) between these three ‘attitudes’ to what the problem is, which then reflects treatment service systems.

View PowerPoint presentation.

Switzerland: Thank you so much. Now Q&A

UNODC – Research: So what is your stance, was the precursors control a success? In the US, it has shown to drive down prices and prevalence of use.

Peter: Production is affected by separately driven variables. The dynamics are that precursor control leads to higher prices and leads to reduction in production in Colombia. At that stage Colombia exclusively produced for the US market and the US exclusively received cocaine from Colombia.

John: In 2006, USA was the largest market for cocaine and no one expected that to change but it did in just 4-5 years – why? I have no idea. We have speculations, it is plausible that precursors were key and it is plausible production sites were key but it’s deeply depressing that the collective community failed to seriously investigate this phenomena. Understanding suppliers matter… yet we under-invest in understanding large scale changes.

Switzerland: It’s interesting that as researchers form the US, you don’t have a clue… what does it say for the policy makers.

Peter: Many areas exist in which evidence base for policy is weak but I guess policies have to be made… The moral concerns are so prominent that we don’t have a great motivation in collecting more evidence. I think that’s changing however, I have optimism and some countries have made serious attempts, but there is not a huge demand for policy research in this area.

Anita: With the cannabis experiments in US states, we found that the increase in research interest can generate better understanding of markets – that is good news.

INCB: precursory events…

John: yes the last one was in 2006

INCB: So the main importer in Colombia was a close friend of the president who died in a funny airplane crash – one big mystery.

Secretary of INCB: About the fentanyl drive: Since it is such a cost effective substance, why does it need to be cut? If u factor in the relative costs, there is still a big advantage?

John: yes. Fentanyl is much cheaper that heroin but it is more expensive than other white powders that you can fill a plastic bag with. That’s is why they are cut. It’s a good way to think about the operation.

USA (health and human services): Chart to cdc: we are concerned about the spike in fentanyl, meth and heroin is also increasing, are you looking into that and possible correlations?

John: Yes, but we think meth is a separate story (price is an important part of it) but heroin fatalities might be largely because of fentanyl and other adulterations. It is complicated though, opioids today are different-recipe of different molecules. There are new opioid antagonists that are good substitutes for each other. I hope in a year we will have better details.

Peter: We think of fentanyl as an opioid, screwing up the market, but it is now mixed in to cocaine as well. Some people deliberately buy a mix of stimulant and narcotic so … overdoses on cocaine are mostly on powders mixed in with fentanyl.

John: These two things, cannabis becoming legal and fentanyl’s appearance, are game changers and can cause unforeseen ripples down the road and not all might be bad. For example, pulling nicotine addicts toward vaporizing or, on the fentanyl side, it’s not entirely crazy to imagine that cocaine users will be repulsed by the reputation of fentanyl in their drugs. I agree that we mostly can be pessimistic, but I want us all to be humble in making predictions.

Switzerland: How would you say liberalization of cannabis laws […] could affect the alcohol market?

Anita: Interesting angle […] Alcohol is involved in some many overdoses. There is no difference in the therapeutic community between alcohol and other drugs. The medications are slightly different but they are basically same in the treatment space.

John: In the economic sense, alcohol consumption patterns changes a lot by cannabis’ availability, alcohol causes a lot of harm. The literature is split in the middle – half the studies say cannabis regulations drive alcohol use down and there are just as many studies saying the other way. I guess we will know in 20 yrs. which is about 19 years too late.

Question: Based on what you differentiated between intensive use and use?

John: Literature is mostly about […] I like this question, you are recognizing the effect on prevalence can go one way if the effect on problematic use goes the other way.  A lot of people believe in cannabis as a medicine and a lot of people believe it is effective in treating a number of conditions and addiction. Under the treaties it wasn’t identified as such, but big corporations also believe that so people in this building will have to grapple with this.

Question: Hopefully the WHO will help us with that…

(Angela Me) UNODC: You would be great panelists, evidence is super important. There is a lot of advocacy out there, dressed as research and not many scholars have as much rigor as I’ve seen today.

Leave a Reply

Your email address will not be published. Required fields are marked *