Voltonomics: A Detailed Analysis of the Chevy Volt's Profitability

UPDATE: LINKED BY POWERLINE! THANKS GUYS!
UPDATE #2: LINKED BY POINTS AND FIGURES! THANKS JEFF!
UPDATE #3: LINKED BY THE VOLOKH CONSPIRACY! THANKS TODD!
UPDATE #4: RESPONSE FROM GENERAL MOTORS (SEE BOTTOM)
UPDATE #5: INSTALANCHE! THANKS GLENN!
UPDATE #6: LINKED BY DOUG ROSS! THANKS DOUG!
UPDATE #7: LINKED BY TOM NELSON! THANKS TOM!

Like what you see? Follow me on Twitter

This post will be the first in a series in which I attempt to bring some clarity to the profitability - or, more accurately, lack thereof - of the Volt.

There is a lot of misinformation floating around regarding the Chevrolet Volt.

Recently reporters have unearthed some very useful data about the economics of the Volt. Unfortunately, these reporters clearly lack the basic financial and accounting knowledge to turn that data into useful information, and instead have muddied the waters with foolish claims, such as Reuters' assertion that Chevy loses $49,000 on every Volt it makes. I explain why this claim is faulty in this post.

For this post, I have modeled out the current and prospective profitability of the Volt, using publicly available data and estimates, as well as some estimates of my own. My estimates are certainly open to debate, but on balance I believe that I have erred on the side of generosity to GM.

THE BOTTOM LINE: even with generous assumptions, the first generation of the Chevrolet Volt will consume about $1 billion in federal tax credits, and STILL result in an economic loss to GM shareholders in excess of $600 million over its lifetime.Without the subsidies, the cumulative loss would triple to $1.8 billion.

Let's start with a look at the Volt's cost of production.

Reuters published estimates of the Volt's marginal cost of production ranging from $20,000 to $32,000. That's an awfully wide range. For the purposes of this exercise, I will use the midpoint: $26,000.

At that cost of production, the marginal Volt is quite profitable. The MSRP is $39.995. Of course, even if someone pays the full sticker price, GM doesn't see the entire $39,995. Dealers have to eat, too. According to Bob Lutz, who ought to know, GM sees a dealer net price of about $37,000, implying an $11,000 profit margin on each marginal Volt sold.

That's nice work if you can get it, but there are some mitigating factors.


Margins Swollen By Subsidies: The first factor is that the buyer isn't actually paying the $39,995 MSRP because he's getting a $7,500 federal tax credit. In other words, fully 68% of the profit on the marginal Volt is the result not of GM's creation of economic value, but of GM's rent-seeking. The money is still green, but it suggests that the volume of Volt sales, weak as it is, is artificially propped up. Absent the tax credit, which expires after 200,000 qualifying sales, GM either sells far fewer Volts at the same profit margin, or the same number of Volts at a much lower profit margin.

Model Year-End Discounting: The second is that GM is actually employing some pretty significant incentives - reportedly about $10,000 per vehicle, mainly in the form of highly subsidized leases. This is probably the driving force behind the Volt's record sales volume in September. This isn't the end of the world; it's just the end of the model year. Presumably GM is moving aggressively to clear out the 2012 models, and will dial back the incentives once the 2012 models are gone. And if their marginal production cost really is around $26,000, then they are still eking out a modest profit on those heavily subsidized leases.

The one caveat is that leasing entails making assumptions about the Volt's residual value, even though there is little data on which to base those estimates, so the profits GM prints today could be illusory, and wind up being offset by writedowns when those 2-year leases begin expiring in 2014.

Low Margins On Fleet Sales: The third is that some proportion of Volt sales are to fleet buyers at substantially reduced margins. I haven't seen a lot of data on fleet sales. In fact, I could only come up with three data points from various media reports. Fleet sales accounted for 35% of sales in December 2011, 7% in March 2012, and 3% in June 2012. The 35% is certainly an outlier. The mid-point of the other two data points, or 5%, seems like a conservative placeholder in the absence of better data.

On the other hand, anecdotal evidence suggests that unless retail Volt sales significantly accelerate, the proportion of lower-margin fleet sales will rise. General Electric alone has said that it will purchase 12,000 Volts by 2015, and the Department of Defense is buying 1,500.

Despite this, in the model I assume that 95% of Volt sales are at the MSRP and earn GM $11,000, and 5% are fleet sales at $30,000, yielding a profit of $4,000. This gives us an estimated weighted average profit margin of $10,650 per vehicle, before accounting for development costs.

Development Costs: Reuters reported that GM spent approximately $1.2 billion bringing the Volt to market. If true, then at an average profit of $10,650 per vehicle, GM just needs to sell 112,676 Volts (total development costs divided by average unit profitability) over the next few years to break even, right?

Not so fast.

Capital Costs: That $1.2 billion spent to bring the Volt to market didn't just fall from the sky. To the extent it was borrowed, GM incurred interest expenses. And any money that wasn't borrowed belonged to GM's shareholders, and if GM is going to invest it on their behalf, it needs to earn enough of a return to make the risks worthwhile. Otherwise, GM should just return the money to shareholders, who can spend it, or invest it elsewhere, as they see fit.

The question of how much of a return shareholders require depends on things like the mix of debt and equity used, and how risky the investment is. There is no single, definite answer, and such discussions can quickly become very abstract, so in the interests of caution and simplicity I will assume a cost of capital of only 10%.

I will also make the generous assumption that the $1.2 billion was invested in equal $400 million increments in 2008, 2009, and 2010. In reality, some meaningful chunk of that money was spent in prior years - the Volt concept car was introduced in 2007 - and so the real-world compounded required return may well be higher than in my model.

Sales Volume: Finally, the big question is, how many Volts can GM sell? Ultimately, if they sell enough of them, they will earn back their investment, including the cost of capital, and can declare victory, albeit a hugely subsidized one.

According to publicly available data, GM sold 326 Volts in the U.S. in December 2010, 7,671 in 2011, and 16,348 through September 2012. The surge in 2012 is a bit misleading since 35% of the year-to-date sales came in August and September, concurrent with those margin-destroying incentives.

Foreign Sales Count, Too: On the other hand, Volt sales are not limited to the U.S. The Chevy Volt is sold as the Vauxhall Ampera in the UK, the Opel Ampera in continental Europe, and the Holden Volt in Australia and New Zealand. In the first five months of 2012 - the only data I could find - GM sold 2,284 Amperas, equal to 32% of sales in the U.S. during the same period.

I have no idea about subsidies, or the retail/fleet breakdown, or overall profitability of overseas sales, so let's err on the side of generosity, and assume that GM can sell one Volt somewhere overseas for every three it sells here, at the same pre-subsidy profit margin (despite the very real possibility that total auto sales in Europe may be more likely to collapse then grow).

For the purposes of this exercise, I assume that GM sells 2,000 Volts per month for the remainder of 2012 (lower than the August-September rate due to the presumed termination of model year-end incentives), rising to 2,500 per month in 2013, 3,000 per month in 2014 and 2015.

In other words, I assume no decline in first generation Volt sales even as the second-generation Volt starts coming to market in 2015. This again errs on the side of caution in judging the continued marketability of first-generation technology in a highly competitive market.

I concede that it is entirely possible that the Volt suddenly takes off, and GM sells many more Volts than I have modeled here. Or lots of the forthcoming Cadillac ELR Hydrid, which I have left out entirely although it ostensibly will further monetize GM's investment in the Volt's technology. But I am aware of no evidence to date that would support such speculation.

It is also quite possible that non-fleet, non-incentivized sales flatline. This uncertainty is why I have tried to give GM the benefit of the doubt with respect to various estimates. As noted above, one could quite reasonably assume a higher required rate of return on capital, lower sales and/or margins, and wind up with losses greater by an order of magnitude than what I have estimated.

Note that I also assume away tax effects. This is another act of generosity, since even if (as I expect) the Volt fails to turn an economic profit, it could still turn a nominal profit and incur tax liabilities.

Finally, this all assumes that there are no recessions or economic shocks over the remaining life of the first-generation Volt that could cause demand to collapse.

With all that said, here is the analysis. I lack the skills to insert an Excel table here (if someone knows how I'd be grateful if they'd let me know), so I'll just describe the numbers year-by year (all amounts pre-tax):

2008: $400 million in development costs, and the 10% cost of that capital, produce both an annual and cumulative loss of $440 million.

2009: Another $400 million in development costs, and another 10% charge for the cost of invested capital, produce an annual loss of $484 million and a cumulative loss of $924 million.

2010: Another $400 million in development costs, and another 10% charge for the cost of invested capital, along with an estimated $3.9 million in profit on the 326 Volts sold, leads to a $528 million annual loss, and $1.45 billion cumulative loss.

2011: No more development costs (we've already accounted for the full $1.2 billion in prior years), but a charge of $145 million on the total invested capital, against an estimated profit of $81.7 million on 7,671 Volts sold nets out to a $64 million annual loss and $1.52 billion cumulative loss.

2012: The Volt's first profitable year, as $152 million in capital costs are offset by $198 million in profit on 22,348 Volts sold in the US and $23 million in profit on 7,375 Volts sold abroad. Annual profit of $70 million reduces cumulative loss to $1.45 billion. One caveat: I have reduced domestic profits by $40 million ($20 million in August and $20 million in September) to account for the profits lost to the high model year-end incentives.

2013: $145 million in capital costs are offset by $320 million in profits on 30,000 Volts sold in the US, and $31 million in profits on 9,900 Volts sold abroad. Annual profit of $206 million reduces cumulative loss to $1.24 billion.

2014: $124 million in capital costs are offset by $383 million in profit on 36,000 Volts sold in the U.S., and $37 million in profit on 11,880 Volts sold abroad. Annual profit of $297 million reduces cumulative loss to $944 million.

2015: $94 million in capital costs are offset by $383 million in profit on 36,000 Volts sold in the U.S., and $37 million in profit on 11,880 Volts sold abroad. Annual profit of $326 million reduces cumulative loss to $617 million.

A $600 million loss after subsidies: So after accounting for all the costs, and trying to err on the side of caution with estimates, the Chevy Volt would have cost GM shareholders $617 million, and U.S. taxpayers just under $1 billion. And with less generous but eminently plausible assumptions about the cost of capital, and sales volumes and margins, the loss to GM shareholders could easily top $1.5 billion.

Without subsidies, the loss would be $1.8 billion: Just for fun, I ran the numbers on what the loss to GM shareholders would be without the $7,500 tax subsidy on each sale. Leaving all other estimates the same, the cumulative loss at year-end 2015 roughly triples, to $1.8 billion.

In subsequent posts, I will examine the economics of future generations of the Volt, and address some of the red herrings offered up by the Volt's defenders.

UPDATE: RESPONSE FROM GENERAL MOTORS

In the comments section at plugincars.com, Rob Peterson of Chevrolet Communications posted a response to the analysis above. He notes that this post's focus on the first generation Volt offers an incomplete picture (which is of course true; I have broken up the analysis into multiple posts because it was just getting too long to expect people to read), and raises some other valid issues. I have invited him to contact me, in the interest of making this analysis as accurate as possible. I hope he takes me up on it. Whether or not he does, in the interest of fairness, I am adding his comments in their entirety:

Blutarsky,

This is Rob Peterson from Chevrolet Communications.

Your analysis, while much more thoughtful, suffers from the same short-sighted assumptions made by Reuters.

First, conservative guesstimates of limited publicly available guesstimates does not make the information any more accurate, it only makes the analysis different. Second, as others have pointed out, the benefits of technologies developed are not limited to the first generation - they will benefit the second and third generations as well.

Like other analysis the fact that several technologies developed for the Volt have already found their way into other vehicles in GM's portfolio is ignored. Technologies such as the battery management systems for the Buick LaCrosse E-Assist and Malibu Eco, electric motors in the forthcoming Spark EV, low-rolling resistant tires on the Cruze Eco and the entire powertrain into the Cadillac ELR (which you did mention) were derived from the Volt.

Also ignored is the halo affect of being the leader in technology. Roughly 60 percent of vehicles traded in for a Volt are non-GM. Your analysis excludes the long-term benefits of attracting new customers who join the ranks of the most satisfied owners in the industry (based on Consumer Reports most recent owners satisfaction survey).

Cost and ROI calculations for automobiles are complex and highly competitive. They must account for tangible and intangible costs and benefits and reflect the impact to the entire portfolio over many years. You only need to look at the impact of the Prius to Toyota portfolio (a vehicle who's humble beginnings mirror those of the Volt) to completely understand that there is more than meets the eyes when introducing an all-new technology.

Thanks
Rob