Should we Delay our Solar System and Wait for More Efficiency, Lower Prices?

Anybody who has bought a computer or a flat screen TV or a smart phone in recent years will understand the emotion behind this article.   If only we had waited…

Fortunately today’s purchasers of solar photovoltaic (PV) systems can put aside their cognitive dissonance.  For PV buyers the most important metrics are efficiency and price.

Efficiency.  Efficiency is the rate at which silicon wafers convert sunlight into electricity.  Today that rate hovers around 20%, i.e., only 20% of the sun’s power is converted into direct current.

(Note:  In this article we vastly oversimplify the nature of efficiency which involves a number of separate variables including thermodynamic efficiency, reflectance efficiency and conductive efficiency, among other things.)

Consumers may rightly ask:  If we wait a couple years will the efficiency of panels on the market increase so that my delay will be justified by greater electricity production?

In contrast to Moore’s law in the computer world (microprocessor speeds will double every two years), solar PV panels have been increasing in efficiency by a mere .5% per annum for years.  That is 1/400th the rate of improvement in computers.

Interestingly, efficiency actually went backwards for a time in the solar industry.  In the 1990s polycrystalline solar wafers were introduced to replace monocrystalline.  The reason was cost:  More polycrystalline could be grown more quickly and cheaply and therefore efficiency was traded for price.  Recently some manufacturers have returned to monocrystalline but efficiency overall remains generally flat.

In any event, the National Renewable Energy Laboratory has followed efficiency developments for years.

cell efficiencies

There have been some breakthroughs and theoretical efficiency levels are quite high.  The EE Times Europe reported in December 2014 that a French-German team achieved over 46% efficiency in a laboratory.   Meantime the MIT technology review reported in June 2014 that Panasonic had broken the historic 25% efficiency mark with one of its panels.

But note that these exciting developments involve very expensive laboratory applications.  For example, in the case of Panasonic the company had to add additional silicon on the front and back of the cell and redesign it to reduce shadowing present in conventional field conditions.

So we are comfortable with the 20% mark for many years to come.

Another way of looking at efficiency is to ask whether new and more efficient solar technologies may come along.  After all, the standard silicon panel has been around since the 1970s.  Surely there must be something new that will be better and more efficiency.

True, there are several alternatives to the standard silicon PV cell.  Cadmium-telluride panels have been used by thin film solar manufacturers such as First Solar.  While these panels produce electricity in more indirect sunlight than silicon they have nevertheless faced headwinds.  A number of thin-film manufacturers have gone bankrupt in recent years such as Unisolar.

Another exciting new development is a perovskite structure that uses an organic-inorganic lead or tin-halide material.  Perovskite solar cells were introduced in 2009 and have already gone from 3.8% to over 20% efficiency with materials that are economically priced and easy to produce.

Given the sharp drop in prices for silicon panels – now running around $.60/watt – it is unlikely that delays over less than ten years could possibly result in meaningful improvement in cost-effectiveness of solar panels.  Consumers will recall that in the intervening ten years they will continue to be exposed to conventional electricity supply costs and volatility.

Pricing.  Another factor in weighing delay is price.

Prices have generally declined steadily over the past 40 years since silicon panels were first produced and marketed.  Bloomberg has published this chart which summarizes the steady trend in silicon PV panels.

historic pv prices

Note that $.30/watt per silicon cell does not reflect the cost of incorporating that cell in a solar panel.  Panel prices have likewise decreased but today are no less than $.70/watt for new panels.  Vintage 2012 panels can be purchased for less than $.50/watt.

This price trend appears to be inexorable.  But wait:  Note what happened in 2005.  In that year Germany ramped up demand for solar and prices actually went up.  The decline after 2008 can be attributed to international economic weakness.  But future lower prices are by no means certain.

Moreover, the percentage savings from continued lower prices is negligible now that solar panels have dropped below $1/watt.  With total installation costs running about $2.20/watt on average, a $.10 drop in panel prices is less than 5% of the total cost of installation.


In some industries delay can be rewarded.  In the case of solar panels the case for delay is harder to justify.  Continuing utility costs as well as volatility could easily offset any benefits that might be realized from greater efficiency and lower prices.