Just to Clarify

A January Surprise

January is a tricky month for forecasting.

First off, most years there are year-end sales events in December.  These events wrap up New Years Day weekend.  This year, this was January 3^rd.  Not surprisingly, sales over this weekend get reported in December to take advantage of the special offers. 

The result is that this January effectively consists of 28 days instead of 31—at least from a sales perspective.

Because of the year-end sales events, the first couple weeks in January tend to be slow—after all, if you wanted a new vehicle, why wouldn’t you buy in December when it was on sale?

Much of this is baked into the SAAR algorithm, which assumes that January will be the slowest month of the year in terms of unit sales.

So the data that is starting to come in is at least a little surprising.  Shopping activity on the web site is “strong.”  (I put this in quotes, because this is a relative thing.  Strong as compared to activity over the past couple years.) 

Add to this the sales data we are seeing—which also looks robust—at least for January.  And remember, this is before the 15^th, when sales typically start to pick up.

So here are our best guesses for January:

December momentum is continuing.  The data points to retail sales of 706k, or a SAAR of 10.62m. (Compared to 949k or 10.36m SAAR in December).   If we assume the same fleet mix (17%) as December, we would see total sales of 851k, or 12.8m SAAR.  (Up from 12.5m)

Retail-share-wise, compared to the same period last month, BMW is down about 10%, while Hyundai and Kia are up about 8%.  Everyone else is relatively stable.

Note how the overall sales are much lower, but the SAAR is actually higher than December.  As I said, SAAR assumes a slow January.

What will be really interesting is the next 15 days.  Will we see a typical sales uptick?  Will this be enough to push SAAR over 13m overall?

Don’t know.  What is clear is that consumers are out shopping and buying.  Not a bad way to start the year.

Our Official October Forecast

Last week I was feeling pretty good about October sales.  This week, not so much.

Sales over the past seven days have slowed considerably.  There is not much here to go on, but this suggests that the higher sales pace earlier in the month was fed by demand for 2011 models--demand that was quickly met. 

We expect the SAAR for October to end up around 11.9M.  This is still a comparatively good result and hints at consumers being a bit more active than we have seen over the past year.  Incentives are also down, which certainly helps automaker profitability.

Retail sales look to be 754K.  (SAAR of 9.7m)  Total sales will be around 920k-assuming 18% fleet.

 

 

 

 

September Sales Forecast

Our forecast is being released today and is in line with what I covered earlier in the month.  Sales started out nicely with bargain hunters out right before and during the Labor Day Weekend events.  The week after, sales activity dropped 40% and has been recovering since.  Looks like the month will end up between 11.4 and 11.5 SAAR with fleet sales being up slightly from August.

The more meaningful Retail SAAR look to be slightly under 9.5, which is about the same as August.

I would say we are still on track for the year to end up around 11.3 sales.

August Sales Forecast

Our "official" forecast for August sales is being released tomorrow.  Not much has changed since I wrote about this a week back--still seeing industry sales coming in at around 11.8m SAAR.  This is a bit higher than July and the highest sales level this year.  Still it is disappointing.  After all, this is the year-end clearance season and the deals have been getting better.  But apparently they are not good enough to entice recession weary consumers into showrooms.  (I know we are not technically in a recession but try explaining that to consumers.)

Incidentally, we are seeing fleet sales at about the same level as last month.   Next month we will start issuing separate SAAR forecasts for fleet and retail.  (I mention a while back why including fleet in the reported monthly sales muddied the waters when it came to inferring consumers demand.)

To break these numbers out, we have been conducting a historical analysis of the correlations between weekly consumer activities and monthly sales.  (We use these correlations as the basis for our forecasts.  Bet you though we were just guessing, didn't you?)  I will probably post more on this later, but I was struck by the percentage of fleet business some manufacturers reported earlier this year.  For example, in February Chrysler's fleet sales exceeded 55% of their total reported sales. 

Sergio gets upset when this gets pointed out, but to me shows how Chrysler has really gone to school on their numbers.  The incremental cost of consumer incentives is huge--in the tens of thousands of dollars.  When you are treading water waiting for new product to arrive, meeting production needs through fleet sales vs. offering big incentives to consumers looks to be a smart move.

One last note on August sales.  GM's sales/share looks to have dropped.  Not great news when you are in the middle of an IPO.

One reason for the decline is inventory--or more precisely the lack thereof.  My understanding is that GM's sales are balanced 50/50 between 2010 and 2011 models.  This compares the industry ratio  of 80/20.  

Getting inventories helps keep costs in line, but it also can result in lost sales.