Model 3 Backlog Conversion Ratio At 30% Challenges Tesla's Growth Narrative

Summary

• Tesla is now taking open orders in North America.
• We estimate a conversion ratio of 30% between backlog reservations and orders.
• At that conversion ratio, the current 420,000 backlog translates into 126,000 deliveries and only 36,000 vehicles at the current ASP.
• The weak conversion ratio challenges Tesla's growth narrative.
• It supports our view that Model 3 annual demand will be capped at 300,000.

Last November, we published an article estimating the demand for Tesla (NASDAQ:TSLA) Model 3 (and Model Y) at various price points. We expected about 280,000 of annual demand for Model 3 priced at USD40,000 and above.

For this analysis, we looked at the price/volume mix of Audi (a subsidiary of Volkswagen (VLKAY)), BMW (OTCPK:BMWYY) and Mercedes (OTCPK:DDAIF) and estimated what percentage of cars was sold in USD10,000 price buckets. We then translated that for Tesla, which was then only active in the higher segments with the Model S and Model X.

The numbers are in Table #1 below; we see a demand of up to 476,000 (for that you would need to sell cars in the USD20,000 area) and 280,000 at USD40,000 and above.

We note the high price elasticity: the demand would be 2.3 times higher for a price variation of USD20,000 (60-70 to 40-50 is 2.5 times and 50-70 to 50 and below is 2.2 times). This is similar to the 2.1 estimate indicated in a GS research note (behind a paywall).

Translating this analysis to the Backlog

If we assume that the distribution of the demand in the backlog is similar, then we can split the backlog into the same price buckets. We also separate the backlog between North America and the Rest of the World using a 50 / 50 distribution. The numbers are in Table #2 below.

Using a current ASP of USD52,000, or 45,000 net of the tax credit (Source: broker estimate), we estimate that the backlog in this price range is 204,233 on the 455,000 backlog, of which 102,116 in North America.

Model 3 backlog conversion ratio is around 30%

As orders are now open in North America, it suggests that the backlog has been exhausted in this price range.

As such, our estimated 102,116 backlogs (in this price range for North America) has been converted into 36,000 sales (28,000 deliveries, to which we add some cars in transit and an allocation for demand in the coming weeks). This gives us a 35% conversion ratio.

However, we cannot use this conversion ratio on the current backlog of 420,000, particularly as many customers no longer need to make reservations to place an order. We also need to take into account reservations and cancellations.

The backlog number of 455,000 dates from a year ago, and during that period new reservations and cancellations have come in. We need to account for these cancellations in the conversion ratio. I make the following simple adjustment. As we forecast a maximum 300,000 of potential demand for Model 3, I think 150,000 of new reservations added in the last year is a conservative estimate (Second Measure has published a more detailed analysis on this). This higher backlog number (605,000) translates into a 27% conversion ratio. For calculation purposes after that, we will use a 30% conversion ratio.

Note that if you don’t take this effect (new reservations/cancellations) into account, we get to a situation where the reduction from 455,000 backlog to 420,000 is more than 100% explained by the deliveries (~100% conversion ratio). While it sounds nice, it is not. It implies that clients wanting a car in the USD45,000 and above price range represented only 20% of all backlog reservations (as orders are now open to new clients for this price range), and thus, 80% of the demand is for a product that does not exist (or is not profitable for Tesla).

An alternative way to calculate the conversion ratio is to take this assumed new total order backlog (605,000) and the current backlog (420,000), which then translates into a conversion ratio of 19% (36,000 divided by the reduction from 605,000 to 420,000). The GS note I mentioned earlier estimates the conversion ratio to be 23%.

Some perspective on our calculation

We make many assumptions in this analysis. First, our two central assumptions are that both Tesla Model 3 demand and backlog will both follow the sales/price mix of other car manufacturers in their price range and a 50/50 geographical distribution. Secondly, there might be material differences in the conversion rate between geography: Europeans may have a higher or lower conversion rate, for example. Finally, as the backlog ages, the conversion ratio may trend lower. A variety of reasons could explain this: Tesla went for existing customers first, customers buying higher-cost vehicles may have less price sensitivity/cash need and are more likely to order if they want to...).

30% may sound low to the most bullish investors. However, it is a fantastic feat to get so many deposits converted into orders. I think this continues to demonstrate the appeal of the product and the Tesla brand.

Model 3 demand is weaker than implied by Tesla valuation

Based on our study, and even taking into account its limits, the existing backlog does not provide a large pool of orders for the Model 3. Using the 30% conversion ratio and the current backlog of 420,000 units, we estimate:

• 124,000 of demand at any price point, which amounts to 25 weeks of production on a 5,000 rate.
• 36k at USD45,000 and above from international customers, or 7 weeks on a 5,000 production rate.
• This explains the rationale for opening orders more broadly.

To be clear, this is not my view on the total demand for the Model 3: backlog reservations are only part of the demand. However, this low conversion ratio and static/reducing backlog support our view that Model 3 demand will be capped at 300,000 a year. As per our demand analysis, we expect to see ongoing demand from North American customers in this price range who have not placed a reservation. We see this ongoing demand being capped at 102,000 per annum (see table 2). On the current ASP, adding the international backlog (36,000) and North America ongoing demand (51,000 for two quarters) points to 87,000 of demand for Q3 and Q4. From a production perspective, this may be challenging enough.

The "true" demand for the Model 3 and Tesla long-term growth potential for Tesla might finally be tested in Q4.

• If Model 3 production gets above 87,000 (cumulative Q3 and Q4)
  • And Tesla manages to maintain the ASP, this would suggest equal or greater demand than my 280,000 per annum estimate.
  • If ASP falls gradually, then medium-term demand is likely lower.

• If Model 3 production is below 87,000, then a stable ASP does not provide us with much information, but a lower ASP would point to weaker demand, as the appetite for the expensive models is less than we anticipate.
• The above test is against my forecast. Tesla valuation implies a much greater demand level.

Conclusion

Our analysis suggests that the effective demand from the current backlog would only absorb seven more weeks of production for cars at the current ASP and 25 weeks at any price point. This is based on a 5,000 weekly production rate. As such, we find the market too optimistic about the strength of the backlog and the Model 3 demand.

Disclaimer: I am not a registered investment advisor, and this article is not advice to buy or sell stock or options in any company. The investor needs to do his own independent investigation that includes reading the company governmental filings and press releases, as well as anything else relevant to determining if this company fits the investor's risk

Disclosure: I am/we are short TSLA. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.