Recreational Marijuana Sales (Graph of the Week)


The big news this morning is that the federal government, led by Attorney General Jeff Sessions is set to rescind the so-called Cole memo, or memos as the case may be. For those who don’t know, the New York Times describes it as “an Obama-era policy of discouraging federal prosecutors from bringing charges of marijuana-related crimes in states that had legalized sales of the drug.” In practice this allowed Colorado and Washington first, followed by Oregon, Alaska, Nevada, and now California to establish recreational marijuana markets and not worry too heavily about federal prosecutors cracking down on these operations which were legal at the state level, but never legal at the federal level.

Last year, our office was tasked with forecasting recreational marijuana tax revenues for the state. You can read our summary here. Included in our work, and discussed with our advisory group was the possibility of changes in federal policy, or direction. This was especially salient given the changes in the executive branch. Our forecast summarizes it as a risk (PDF pg 38):

Finally, the one risk that looms large over the entire forecast is the federal government. While there has been no clear warning or action taken, there is a non-zero chance the federal government could step in and eliminate, or severely restrict recreational marijuana sales. In this event, taxes collected would be considerably less than forecasted.

Well, now there does appear to some action taken. Ultimately it will likely come down to enforcement, and the choices prosecutors make. We will be meeting with our advisory group again this month, and will discuss the implications of these changes, along with other issues and trends in the recreational marijuana market. More on this after our meeting and as we get closer to the Feb 16th forecast.

All of this brings us to this edition of the Graph of the Week, which shows monthly recreational marijuana sales for Colorado, Washington, Oregon and Nevada. These sales figures are estimates based on reported tax collections, and do not include medical marijuana sales. In total across the four states, they are seeing around $250 million in recreational marijuana sales per month. Additionally, all states continue to see growth in this newly legalized world. However the exact level of sales is also determined by the size of the population, usage rates, and tax policy, among other factors.

In fact, in my preferred chart in comparing sales trends, Oregon’s first year and a half of sales are nearly identical to Colorado’s first year and a half of sales, after you control for population size. Let’s call this the Bonus Graph of the Week. Also note that Nevada is seeing strong sales in their first few months. Nevada is currently selling about the same amount of recreational marijuana as Oregon is today, however with a much smaller population. Their initial adjusted sales data is the highest we’ve seen among the legalized states. Nevada and Las Vegas in particular are also a tourism hub, and thus are seeing larger sales than the resident population alone would suggest. I don’t have a huge reason to believe cannabis tourism is a big factor in Oregon’s sales, but I think it clearly is in Nevada.

As we write in our forecasts, there are a lot of risks to the recreational marijuana outlook. In particular, usage rates, prices, harvest levels, regulations and the like both have upside and downside implications for the forecast. However, none of those loom as large as changes in federal policy. Today’s actions may end up mattering substantially, or not so much, but we don’t know the answer yet. Our office will continue to work with our advisors, and to adjust the tax revenue outlook accordingly.

Oregon Recreational Marijuana Forecast


SB 845, among other things, would give our office the recreational marijuana forecast responsibility. While not current law yet, we went ahead and produced such a forecast for the first time in our most recent quarterly forecast. What follows below the fold is an extended summary of our forecast work, including lots of pictures, I mean charts, for those interested.

Currently the outlook for recreational marijuana sales and tax collections remains highly uncertain. While Oregon has now collected just over a year’s worth of taxes, there have been substantial changes during this time that complicate any analysis. Early start sales through medical dispensaries were taxed at a 25 percent of rate, while sales at OLCC licensed retailers are now taxed at a 17 percent rate, with the local option of adding up to 3 additional percent. Furthermore, regulatory changes, more stringent product testing requirements, and Mother Nature all impacted and reduced available supply on the market during this time.

The first chart shows monthly tax collections as reported by the Department of Revenue, with the colors representing the different state tax rates. During the transition period there were recreational sales at both medical dispensaries (taxed at 25%) and at OLCC retailers (taxed at 17%).  The percentages listed are the effective tax rates after blending together sales under both regimes. These figures do not include any of the local option taxes, just the state portion.

As such, it is challenging to get a handle on the underlying trends in this newly legalized world. Thankfully, Oregon is not alone. Both Colorado and Washington are two years ahead of us. Both states have seen tremendous growth in sales and tax collections, which serves as a guide for where Oregon is likely headed in the near-term. Over time, as the market matures, future growth will follow trends in the economy and consumer spending. However the coming few years will see strong growth as the product becomes more widely available, more socially acceptable, and more black and gray market sales are realized in the legal market.

Certainly, one year’s worth of tax collections, and one set of quarterly tax returns filed by dispensaries is more valuable than no data. Our office’s forecasting responsibilities are made considerably easier than what faced those estimating the potential impact of Measure 91 (2014) which legalized recreational sales. That said, one year’s worth of data is not enough to build a full-fledged forecasting model, particularly when it is a brand new legal market. Over time, as we accumulate more data, a longer history of sales, and detailed breakdowns of consumer purchases and demographics, our office will build an econometric model. Until then, in consultation with our advisory group, and using Colorado and Washington as a guide, our office is relying on trends for the short-term outlook.

In terms of sales, Oregon’s first year closely tracks Colorado’s first year and outpaces Washington’s. The numbers in the graph below are estimated sales figures based on actual tax collections. Both Colorado and Washington are larger states than Oregon, so we adjust their figures based on the relative size of the adult population.

There are at least four main reasons for this pattern:

First, marijuana usage rates from surveys indicate a larger share of Oregonians have used marijuana in the past month than what is reported in Washington. As such, Oregon is more likely to see larger sales than Washington, after adjusting for population size. However, usage is not the only measure that matters, as Colorado’s usage rates are even higher than Oregon’s. Note that these national health surveys are based on responses prior to recreational legalization.

Second, prices and taxes matter. Oregon has a significantly lower tax rate than does Washington, which helps keep final consumer prices lower. Furthermore, the first set of quarterly tax returns, a very limited data set, indicates that Oregon prices were very competitive with Washington prices, even though Washington had two additional years to get accustomed to the newly legal market, license growers, processors and the like. A lower retail price, everything else equal, should bring more consumers and more black market conversions.

Third, the cross-border effect with legal sales beginning earlier in Washington likely had an impact on Oregon’s first year of sales. Counties in southwest Washington saw sales fall by nearly 40 percent once Oregon’s early sales began. Clearly there was plenty of cross-border activity. Effectively this meant Oregon had somewhat of a built-in customer base who were used to purchasing in the legal market. Thus Oregon’s initial sales were larger than in Washington, but this may have some to do with social acceptance and being used to the new system rather than fundamentally stronger sales.

Fourth, both Colorado and Washington initially had relatively few retail outlets in major population centers. In Colorado, Denver had retailers but Boulder did not initially. In Washington, Seattle had only a few retailers at first, but have added quite a few in recent years. As such, some of each state’s strong growth in the first two years was simply due to market access and product availability, particularly in places where lots of people live. It is unlikely this is a similar issue in Oregon, with our major population centers having dispensaries at first, and retailers now. Not that Oregon is overstored, or that there cannot be more room for growth – Colorado, for example, has considerably more retailers even after adjusting for their larger population – however lack of consumer access does not appear to be a major issue in Oregon today for much of the population.

In terms of the outlook, Oregon is poised for strong growth in the coming years. However, given the above and the advice from our advisory group, our office is not forecasting revenues to be quite as strong as those seen in Colorado over their second and third years.

This outlook remains highly uncertain with substantial upside and downside risks.

On the downside, supply constraints that keep products and inventory low will result in fewer sales, and tax collections. Such constraints could be regulatory changes that impact grower, processors or retailers, or regulatory bottlenecks where companies in the industry are unable to get their licenses, renewals or tests completed or approved in a timely manner. Another downside risk for tax collections are prices, given Oregon levies the tax based on the sales price. To date in Colorado and Washington, prices have fallen around 20 percent per year. Marijuana is a commodity and eventually will be commoditized. How far and how quickly prices decline is a considerable risk to the outlook for tax collections. Offsetting this risk somewhat is the fact that lower prices should result in larger sales, helping to buoy tax collections overall, which is what has happened in both Colorado and Washington so far. Finally, the one risk that looms large over the entire forecast is the federal government. While there has been no clear warning or action taken, there is a non-zero chance the federal government could step in and eliminate, or severely restrict recreational marijuana sales. In this event, taxes collected would be considerably less than forecasted.

On the upside, consumers overall could get more comfortable with legalized recreational marijuana sales, and the industry gains broader social acceptance, resulting in larger sales. Furthermore, a faster rate of black market conversion would also result in more legal sales. Similarly, conversions from the medical marijuana market to the recreational market would result in more sales and taxes collections. The impact of the seed-to-sale tracking system may also increase activity within the legal market and restrict the flow of product into the black market.

While the sales and tax collection outlook is uncertain, it is also fairly straightforward. The same cannot be said for distributing the taxes, or at least not yet. Currently there have been no distributions from the collected recreational marijuana taxes and there are likely to be none in the current biennium. Start-up costs to OLCC and other state programs need to be repaid first, with only the net revenues after accounting for these costs available for transfer to recipient programs like schools, state police, city and county law enforcement and the like. The exact reimbursement figures will be finalized in the coming months, with the first tax distributions made early in the 2017-19 biennium.

The process and timing for future tax distributions is as follows. First, retailers pay taxes on a monthly basis. These are the figures reported periodically in the media and other outlets. However these taxes are not immediately available for distribution. They only become available for recipient programs once the Department of Revenue has received and processed a retailer’s quarterly tax return. This ensures transfers are made based on the correct, not estimated, taxes paid by retailers. As such there is a time lag of between one and two quarters from when taxes are initially paid to the Department of Revenue and when they are available to transfer to programs. This discrepancy is likely to shorten some in the future as retailers file their taxes in a timelier manner, however the time lag will not be eliminated entirely. The chart below tries to show the lag between the tax collections paid on a monthly basis and when they become available for distribution.

Given no distributions will be made in the current 2015-17 biennium, the accumulated revenues are carried forward into 2017-19 and will be distributed then. Again, note that the Tax Revenues reported in our office’s forecast tables represent the amount available for distribution. Currently there is approximately $78 million in tax collections at the Department of Revenue, with the total 2015-17 figured forecasted to be nearly $90 million. However, as our table shows, under normal circumstances where we do not have to worry about start-up costs being repaid first, only about $67 million of the $90 million would be available for distribution in the 2015-17 biennium. This time lag between monthly collections and quarterly tax returns is a big deal in the budget world and can be a bit confusing. At least for me it was at first when trying to wrap my head around when monies will be distributed.

The above is what our office worked on and published in our latest quarterly forecast. However, some of the costs and revenue distributions will change (or have already changed) based on legislation. For example, the same SB 845 that gives our office the forecasting responsibilities also changes to the revenue distributions. Also, SB 1057 impacts the administrative cost estimates. Additional bills remain alive in the legislative process. Our office, with the help of other agencies and legislative staff, will update the forecast as bills become law. Check back in August when we release our next forecast for all the details.

2015 Outlook: Measure 91


One of the more interesting and yet unknowable questions in the year ahead is what the impact of Measure 91 will be for Oregon. The basics of the vote itself and the direct implications for Oregonians have been well covered. However, the broader and bigger impact on the state’s economy and public resources are not known. The revenue estimates are fairly modest in size — relative to the size of the state budget — and the track record of such estimates in Colorado and Washington are somewhat mixed. That does not mean the impact won’t be felt, as it will. Bringing a largely illegal activity into the legal marketplace will have many benefits including the tax revenue but also additional jobs that will now have legal protections as employees can be covered under the unemployment insurance system for example, plus new (legal) businesses, operations, the regional impact in places like Southern Oregon and the like. Of course, as with all vices, not every aspect is positive and there are some downside risks associated with legalization, even if the research consensus is that marijuana is a safer alternative than other options.

However, from an economist’s perspective the two most interesting aspects are consumer preferences and potentially problematic market distortions. As for consumer tastes, what is the substitution effect of legal marijuana and how will it impact, say, alcohol and tobacco sales. Are these products purely substitutes? To what degree are they complements? University of Oregon professor Ben Hansen gave a very good and informative presentation at the Oregon Economic Forum a couple months back. Among other items, he discussed how not only are other types of products substitutes for marijuana, but how the various markets for marijuana itself can be substitutes, these include both the medical and recreational markets but also personal cultivation.

Along these lines, one interesting but also potentially troubling development is the differential between Colorado’s medical and recreational marijuana markets, due to different tax structures. In essence, due to higher taxes on recreational marijuana, it is less expensive to buy medical marijuana in Colorado. As such the state has not seen a big switch into the recreational market from medical patients and even seen medical sales increase. However, Colorado has seen a switch from the black market into the legal market (be it medical or recreational). That’s the big win from both an economic and societal point of view, bringing the illegal market into the legal one.

However these market distortions between medical and recreational can be problematic not only in terms of rules, costs and regulations but also in terms of achieving the desired outcomes or goals. Consumers will likely figure out the best option for themselves (below is a hypothetical example using Oregon’s medical application fees and Colorado’s price differentials) however that may not be the best outcome for the state overall. That’s why the work the OLCC is currently undergoing is so important — to figure out what those outcomes should be and how to set up the rules and marketplace to best achieve them.


Lastly, one additional tidbit that was somewhat surprising — to me at least — was the overall lack of clear connection between the voting results of Measure 91 and the local medical marijuana population already in the state. Besides the obvious that cultural or societal preferences and ideology has swung significantly in favor of marijuana legalization, regardless of the local medical population, there are a few additional possibilities for these results. One, particularly for a place like Josephine County with a very large OMMP patient population and yet only voted 50-50, could be that the largest medical marijuana market participants did not want to rock the boat, so to speak. They may be relatively content with the existing market and not wanting to shift the landscape for something unknown. Another possibility may be that these two populations aren’t the same. The voting population may be significantly different than the medical marijuana population. Another may be that while OMMP patients are equivalent to about 2 percent of the adult population in Oregon, that may not be a large enough share to influence overall voting patterns. Regardless, these results are interesting to see and the Oregonian has a nice interactive map of voting results across the state, even down to the precinct level in some places.


Stay tuned for a few more posts on the 2015 outlook in the coming weeks, plus a new economic recovery scoreboard that tracks progress across a whole host of measures.

Our office has received a number of comments, questions and requests about the impact of Measure 91 in Oregon, so I thought I would write down a few thoughts and concerns from our perspective heading into the new year. To be clear, so far our office has not been involved in the process at all. As with all initial revenue estimates, those are done by the Legislative Revenue Office (plus there were outside estimates as well) and since the Measure 91 revenues do not go to the General Fund (which is largely our focus) we have not been involved at this point.

Drug-Related Deaths Up 33% in Lane Co.

drugsEUGENE, Ore. — Drug-related deaths were up 33% in Lane County last year, compared to the year before. The Oregon State Medical Examiner’s Office found Lane County saw one of the most notable increases with 20 deaths last year, five more than the year before.

Thirteen of the 20 deaths were meth related, nine blamed on heroin, and two were a combination of drugs.

Doctors say these numbers are alarming, and they don’t show the number of people abusing prescription medications.

“Many of these deaths and these injuries are related to prescription drugs being diverted,” says Doctor Patrick Luedtke, the Chief Medical Officer for Community Health Clinics of Lane County. “Thirty years ago that was not as much of an issue in this country, even ten years ago, and now were have 60-70% of all overdose-deaths related to diverted, legal medications.”

Diversion means you took a prescription drug that wasn’t prescribed to you. He says it’s a big problem, and a major concern among doctors.

High tech medical scans clear way for better local diagnoses


by Sarah Nicholson, EDN

A newer, safer type of medical imaging has made its way to the Eugene area, and those in the medical field are seeing promising returns.

Oregon Imaging Centers recently received a three-year approval for PET/CT imaging from the most highly-regarded accrediting body, the Intersocietal Commission for the Accreditation of Nuclear Medicine Laboratories (ICANL).

PET/CT imaging and scanner from Oregon Imaging Center

Positron Emission Tomography and Computed Tomography, or PET/CT imaging, is a combination of two scans
that allow medical practitioners to pinpoint the location and extent of cancers, dementias and cardiac illnesses by looking at metabolic and structural changes in the body. 

Oregon Imaging Centers is the only center in Oregon approved for both oncological and neurological diagnoses, and only center approved by ICANL, which is considered the gold standard of nuclear medicine accreditation providers.

In 2007 an article published in the New England Journal of Medicine publicized the overuse of CT scans in the US, and the related cancer risk of exposing organs to low doses of radiation. Since then, media reports fraught with information about the dangers of high radiation levels have caused fear in many people about being overexposed.  The FDA also acknowledges that CT scans are associated with a higher risk of lifetime cancer and says the usage of adult-sized radiation doses in children can be particularly harmful.

The combined PET/CT scan delivers less radiation than a regular CT scan, so the danger of radiation is greatly diminishedAnd in the event that cancer is already suspected, the fear of low-dose radiation pales in comparison to other disease-related complications. Additionally, over 90% of the radioactivity from the PET portion of the scan is gone by the time the patient leaves the exam.  The only populations this diagnostic tool is contraindicated for are pregnant and lactating women, but even in lactating women the radiation is thought to be cleared from the breast milk within twelve hours of the test.

An example of (A) CT imaging, (B) PET imaging, and (C) combined PET/CT imaging. From Wikipedia/Renato M.E. Sabbatini, PhD


The PET/CT process uses an injected radioactive substance called fluorodeoxyglucose (or FDG) to highlight abnormal areas within the body. Similar to glucose, which fuels tissues throughout the body, FDG is taken up more eagerly by high-metabolizing abnormalities in the body. So areas that are more metabolically active, like cancerous tumors, draw attention to themselves through their greediness.

Neurological disorders such as Alzheimer’s and Parkinson’s diseases respond similarly in the brain. With an injection of FDG, the compound will accumulate in certain areas, which will then be highlighted on the PET scan images. Add the CT component (Computerized Tomography,) and you have a very specific picture of the disease.

PET(-only) scanners, which Oregon Imaging Centers used before receiving the new accreditation, can differentiate between malignant and benign lesions, and determine the spread of disease. Adding the CT component is incredibly helpful in identifying specific anatomic locations. The dual imaging technique of the PET/CT provides a much more specific picture of disease, especially when it comes to detecting cancer.  

The PET/CT scan is of enormous assistance in both diagnosis of cancer, and in evaluating the effectiveness of therapy.  Luke Breazeal, manager of the PET/CT Department at Oregon Imaging Centers said there are four key times to perform a scan: 

(1) When something malignant is detected or highly suspected
(2) to identify the extent
(3) in the middle of chemotherapy
(4) after completion of chemotherapy 

With just one scan, you can identify cancer, determine its extent in the body, and stage (or esta
blish how far the cancer has progressed.)  For many types of cancer a PET scan can show if a therapy is effective: If a significant decrease in the tumor burden is not seen after a few rounds of chemotherapy, the therapy can be switched, saving valuable time, reducing cost, and increasing life expectancy.

PET image of a brain with Alzheimer's

Although over 90% of PET/CT scans address oncological disorders, it’s an incredibly helpful tool in neurological diagnosis and treatment planning.  In “dementia imaging,” Breazeal said, “we see characteristic metabolic patterns for each specific type of dementia.”   Each of the dozens of different types of dementia displays a unique pattern of glucose (and thus FDG) uptake, and identifying the correct type puts doctors on the right track to choosing a treatment plan. Before PET scans, the only way to definitively diagnose Alzheimer’s was through autopsy.

With the green light from ICANL, Oregon Imaging has seen about a 50% increase in volume for PET/CT imaging.  According to Breazeal, the clinic does about 80-100 scans a month, mostly for oncological purposes.  More information about Oregon Imaging Centers and its new accreditation can be found on its website at

Another example of PET/CT fusion imaging of the whole body



More than Social Media


Seeing a fundraiser or benefit going on for a charity or struggling public entity is almost commonplace. Seeing a fundraiser for someone from your community that you will likely never meet or come in contact with is not. Ever heard of the I-5 Band Coalition? Chances are that’s a no. Recently this local facebook group of “outside the mainstream” musicians and their supporters decided to do something about helping one of their own. In a gesture reminiscent of the barn raisings and harvest parties our grand parents and great grandparents would tell stories about, local promoter Tina Hicks sent out the call to the “neighbors” about the need.

Lawrence Aguayo, a promoter form Myrtle Creek, Oregon, a husband of 22 years, and a father of four, was diagnosed with hepatitis C eight years ago, and is currently battling cancer. Lawrence and his family are similar to a lot of families these days, in that they have no medical insurance. His treatments are estimated to cost somewhere around $9k, all depending on how his body reacts.

After being diagnosed Lawrence was told he would need radiation treatments. Upon arriving at his first appointment he found that he could not receive treatment due to lack of insurance or the ability to make a cash payment. Once the word was out about his needs, his facebook family and freinds began pulling together to help him out. Several fundraiser benefits have been arranged so that we can get Lawrence his treatments as soon as possible. If you would like to help Lawrence and his family out please attend a fundraiser Event. Then next one is scheduled for:

April 29th: Dixie Creek Saloon, 32994 Oregon 99E, Tangent, OR 97389.
Audiophobia, Full Circle, & Kevin Woodring.

Local musicians have come out of the woodwork so to speak to lend their time and abilities to the cause – on their own dimes. Many of the individuals in these bands are only a little bad luck away from being in the same situation as the people they are reaching out to help, yet they don’t think twice about stepping up. This is the kind of thing that gives you a little hope for the world, and the community. We thought this was worth sharing.

If you’d like to help out, you can reach Tina Hicks on Facebook via the I-5 Band Coalition page. EDN is glad to be a part.