The marijuana industry has been practically unstoppable of late -- even in the U.S., where sales are prohibited by federal law, but allowed in 29 states that have legalized weed in some capacity since 1996. According to ArcView, in partnership with BDS Analytics, the average annual sales growth expected between now and 2021 in North America is 28%, leading to a projected $24.5 billion in yearly sales by 2021.

There's absolutely no question that marijuana stocks are growing at an exceptionally fast pace, which has translated into triple- and quadruple-digit percentage gains for many of the largest pot stocks by market cap over the trailing two-year period. However, not all investors are necessarily comfortable owning high-growth stocks.

Some prefer to pad their long-term appreciation potential with dividend income, which begs the following question: Will marijuana stocks ever pay dividends? The answer is more complicated than you probably realize.

A cannabis bud lying atop a messy pile of hundred-dollar bills.

Image source: Getty Images.

Will pot stocks ever pay dividends to shareholders?

While I'd like to say either "yes" or "no," there are far too many variables at play to determine whether the marijuana industry can generate consistent dividend income for investors. But what I can say with relative certainty is four things would need to occur in order for marijuana stocks to consider paying a dividend regularly to shareholders.

1. Cannabis growing costs would have to fall considerably

To begin with, there would have to be a pretty substantial decline in per-gram cannabis growing costs if marijuana stocks are going to pay a dividend. Thankfully, this is a trend we have been witnessing among the largest Canadian growers. Of course, this trend would have to prove sustainable as production ramps up with the expectation of recreational pot legalization by this summer.

A good example is offered by Aurora Cannabis (ACB -0.69%), which is expected to be a top-three grower by total production. Aurora is currently producing just a pittance of the 240,000 kilograms to 270,000 kilograms it could be delivering annually. In the second quarter of its current fiscal year, the company paid a cash growing cost of $1.36 per gram. However, the company's January investor presentation suggested that the higher yields associated with its highly automated Aurora Sky facility and its partner Aurora Nordic facility in Denmark should help bring its growing costs below $0.78 (CA$1) per gram of cannabis. 

This doesn't mean Aurora Cannabis is necessarily going to be paying any dividends anytime soon, but it's the perfect example of the declining-cost model that could portend a sustainable business.

Jars filled with cannabis and stacked on top of each other.

Image source: Getty Images.

2. Supply and demand would have to find harmony

Secondly, we'd have to get some sort of feel for the supply-and-demand picture for the legal cannabis industry.

With Canada on the verge of legalization this summer, expansion from the more than seven dozen licensed producers could generate between 1,500,000 kilograms and 2,000,000 kilograms of dried cannabis production a year by 2020. However, early estimates from various analysts and government reports suggest that annual domestic demand may only tally roughly 800,000 kilograms. In other words, there's the real possibility of a marijuana glut wrecking margins, initially. It's possible that exports to countries where medical cannabis is legal could absorb a portion of this domestic oversupply, but it's unknown exactly how much. 

Considering that no other developed country has legalized recreational weed, there's really no historic-use case to examine. Flying blind can be somewhat dangerous for investors, which makes it even less likely for pot stocks to offer a dividend until there's some sort of stability in the supply-and-demand picture a few years down the road (at the earliest).

3. Additional countries would have to legalize the drug

Next, I strongly believe we'd need to see at least a few other countries legalize marijuana for adult consumption. We've witnessed Europe and a few other countries embrace the use of medical cannabis for select ailments, but the real growth-and-demand story lies on the recreational side of the equation.

My suspicion is that marijuana stocks would have little or no incentive to pay a dividend as long as the bulk of their production is reliant on demand in just a handful of countries. Only when growers feel as if they've adequately diversified their revenue stream geographically will the idea of dividends being paid to shareholders even creep into the picture.

An indoor cannabis grow farm.

Image source: Getty Images.

4. Businesses mature and reinvestment begins to dwindle

Finally, we'd have to witness some form of business maturation. Right now, marijuana stocks are reinvesting their operating cash flow and raising capital faster than you can blink. They're doing this in order to capture market share, pending Canadian adult-use legalization, and because they view this as the best use of capital. When reinvesting in new grow farms no longer becomes the "best use of capital" -- i.e., when the legal weed industry has matured -- it then could be worthwhile for profitable pot stocks to consider instituting a regular quarterly or annual dividend for shareholders.

To be crystal clear, I don't believe we're anywhere near the point of dividends yet in the marijuana industry. In order to get a clear picture of supply and demand, and for new countries to legalize, I'd suspect we're four or five years out, at minimum, if not longer, from even the remotest chance of a regularly instituted quarterly or annual dividend.