Three years after SolarCity acquired high-efficiency solar module maker Silevo, its new owner, Tesla (NASDAQ:TSLA), doesn't appear to be backing down on plans to expand solar manufacturing. It's Buffalo manufacturing facility, which is being run partly by Panasonic, is now reportedly churning out solar cells and will soon begin producing the company's much-anticipated solar modules and solar roof tiles.

Eventually, says CTO JB Straubel, the plant could make 2 GW annually of solar cells for modules and tiles. That would make it one of the largest solar plants in the world, an aspiration that comes with risks in a highly competitive industry.

Rendering of the solar roof on a home with a Tesla car in the garage.

Image source: Tesla.

Tesla has a lot to prove in solar

Most of the reporting on the Buffalo plant so far has highlighted how big it is, and how extensive Tesla's solar manufacturing ambitions are. It's also worth noting that Tesla and Panasonic have a lot to prove before investors can gauge how much credit it deserves for its solar products. 

The Silevo technology acquired in 2014 was all but abandoned when Tesla acquired SolarCity. It was replaced by Panasonic's high-efficiency cell technology, some production of which has been shut down in Japan because it wasn't cost competitive in the market. It's possible that Tesla and Panasonic have found a way to lower costs or improve efficiency, but in the highly competitive solar market, it's not clear that either company can make a competitive solar module at scale. 

The subsidy that makes it all possible

Investors also need to understand that none of this would be possible without a $750 million subsidy from the state of New York. The state owns the building and the manufacturing equipment Panasonic and Tesla are using to make solar cells and modules, which brings both advantages and disadvantages. The rent it's paying to use those government-owned resources is a whopping $1 per year, a price no competitor will be able to beat. 

Early on, the subsidy will be great because it reduces the upfront costs and risks associated with building out solar capacity. But the solar industry is notoriously fast moving, and manufacturing equipment is usually obsolete just a few years after it's built. This is something that's been a constant struggle for commodity solar manufacturers with over 5 GW of capacity. Enterprises like Canadian Solar (NASDAQ:CSIQ) and JinkoSolar (NYSE:JKS). They've struggled to generate a profit because module prices drop and equipment becomes obsolete before it ever pays off. High-efficiency manufacturers like SunPower (NASDAQ:SPWR) have faced a similar challenge, constantly needing to write off old plants and build new ones. And all three are currently upgrading equipment to efficiencies that will likely meet or exceed what Tesla is making and they all likely have the expertise to make solar modules at a lower cost. 

The challenge for Tesla won't be the early production of solar cells; it will be upgrading or replacing equipment three to five years down the road. If the state owns the equipment and Tesla has agreed to rent it, complications will arise over the questions of which parties will pay for what upgrades, and who takes the loss on obsolete equipment. Suddenly, manufacturing capacity may get much more expensive, at which point Tesla will run into the same economic problems that Panasonic had in Japan, and that nearly every solar manufacturer in the U.S. has faced. It's not clear how Tesla will dodge the economic pitfalls that caused the recent bankruptcies of SolarWorld and Suniva

Who is buying all of these solar modules? 

One has to wonder: Who will buy 2 GW of solar modules from Tesla. The company's installation arm is shrinking right now and is on pace to install over 500 MW in 2017 -- well below the 1.25 GW annually that it once targeted. It's unlikely the business will quadruple in size with Tesla de-emphasizing the solar sales channel. 

Given the fact that Tesla is a solar installer, it would be odd for competing installers to buy Tesla solar modules. Why help out a competitor, particularly when Tesla won't be making a product that's significantly (or any) better than other solar modules on the market? That is, unless Tesla sells solar roof tiles to competing installers, something it hasn't given an indication it will do. 

For three years, SolarCity and now Tesla have been talking about their grand aspirations of manufacturing a solar module in the U.S., and for three years, customers and investors have been waiting. It appears that Tesla's manufacturing plant is finally up and running, but turning it into a money-making operation will be harder than it seems. The rubber is hitting the road for Tesla's solar plans, and I have my doubts that it will succeed where so many others have failed before.

Travis Hoium owns shares of SunPower. The Motley Fool owns shares of and recommends Tesla. The Motley Fool has a disclosure policy.