Morgan Stanley is taking a bearish position on the global solar industry amidst tightening regulations and a surfeit of new capacity, and forecasts that wafer and module prices will slide by 10-13 percent in what remains of 2016, and a further 10-15 percent in 2017. Entelligent asked Stephen Byrd, Morgan Stanley’s head of North American equity research for power/utilities and clean tech, what the bank’s downbeat projections mean for clean-energy investors.

How much trouble is the solar industry in?

In our recent global report, we changed our outlook to cautious on the sector overall. What we were struck by, when we looked at the global picture, was the very, very large increase in manufacturing capacity, without a corresponding increase in demand. We actually see 15GW of new manufacturing capacity coming online in 2016, but in 2017 global demand is going to be down from 2016. That’s very concerning. 

Are you seeing problems across the board? 

It’s slowing in a number of countries. In Japan, there’s a pretty dramatic reduction driven by regulatory change: solar incentives there are far less attractive than they were, and that will result in a lot less development. We still think that long-term, China will be a good market, but our analysts see a slowdown relative to the huge growth of prior years. And in the US, there’s a fascinating dynamic: last year there was a huge rush to get things done before the expiration of the federal Investment Tax Credit, but it wound up getting renewed, so now there’s a lot less urgency on the part of bigger customers. 

Why are we seeing this imbalance in supply and demand? 

What concerns me is that I’d like to see more barriers to entry. For any industry, profitability can be brought back to the degree to which there are barriers to entry — but in the solar sector, it’s fairly straightforward for manufacturers, especially in Asia, to simply increase capacity, and there’s not much to choose between their products. 

Could the competition be good for consumers? 

Generally, competition does end up providing low prices to customers, but if there’s enough disruption and the cost of financing goes up — and that’s something we’re starting to see signs of — that would ultimately be a negative for consumers. 

Is this something you’re seeing in other clean-energy spaces? 

We recently put out a briefing comparing barriers to entry for wind vs. solar, and we found the barriers to entry in wind were much higher. The degree of expertise you need to develop wind is higher, and you need access to the best wind resources. Then there’s operations and maintenance: big companies like NextEra (NYSE:NEE) do a lot of their own maintenance in house, so they avoid the third-party fees that smaller companies pay. It’s just much harder for new players to break into wind, which means that returns for established players have been much higher, and we expect that to continue. 

That’s also true of residential solar — companies like SolarCity (NASDAQ:SCTY) and SunRun (NASDAQ:RUN) have an advantage, because scale becomes a barrier to entry. The cost per customer, if you aren’t already a big entity like SolarCity, is going to be a lot higher, and that’s going to make it harder to make a profit. 

How do government policies factor into this? 

Utility-scale solar is becoming less and less policy driven — it’s now much more dependent on the economics. We did an analysis based on First Solar’s (NASDAQ:FSLR) operations in Texas, and we saw that in the not-too-distant future, the required revenue for a large-scale solar facility in Texas would be $40 per MWh. That’s lower than for a combined-cycle asset, and while it does include the federal subsidy, that subsidy will remain in place for quite a while longer. 

Could solar survive without the subsidies? 

In some parts of the country, where the sun conditions are very strong, yes. In others, it’s not likely. If you look at the upper Midwest, where the sun isn’t that strong and utility bills are low, it’s hard to see how solar could compete without a subsidy. In the South and Southwest, there’s a greater possibility of that happening based on the math we’ve been running for large-scale solar. 

You’ve forecasted that electricity demand will flatline — is that bad news for solar? 

Other things being equal, it’d be better to have demand grow — but because solar’s penetration is still so small, we should be able to see very high growth even if demand is completely stagnant. What helps solar is that the cost of solar is falling very rapidly, which gives the sector plenty of economic headroom. SunRun CEO Lynn Jurich says we should be thinking about growth in the 20% range, which is less than some investors were thinking, but is still by any measure quite a good growth rate. 

How can investors pick winners in this sort of environment? 

We’d want to see care with the balance sheet, not getting over extended, a technology advantage in terms of the product they’re manufacturing, and very strong marketing and development capabilities. Those would be the critical things. So, for instance, we’re overweight on SunRun, which has a long track record in distributed solar — the barriers to entry are higher there, and SunRun is very good at marketing and financing. 

Are there any other bright spots? 

For an investor with a long time horizon, First Solar’s technology is promising. We compared First Solar’s thin-film panels to other manufacturers’ technology, and if they can achieve the roadmap they’ve laid out, which they’re making progress towards, then we see a demonstrable economic advantage for their product in the longer term. That advantage really starts to become significant in the later part of this decade — there’s an advantage today, but given the manufacturing overcapacity, the competition is vicious. 

When do you expect things to improve? 

We believe 2017 is going to be a tough year: all the manufacturing capacity will be online, and we’ll also be seeing a pause in demand. In the U.S. and China, we expect demand for solar to pick up later in the decade, although we could be wrong if manufacturers show a lack of discipline and continue to increase capacity. For investors with a long time-horizon, though, there are certainly opportunities to buy cheaply, given where stocks have reset to.

Ben Whitford is the U.S. correspondent for The Ecologist. He has written for the Guardian, Newsweek, Mother Jones, Slate, and many other publications.