EIA: Wrong On Coal Forecasts

|
 |  Includes: ACIIQ, ANRZQ, BTUUQ, CNX
by: Mark Hayes

The Energy Information Administration (EIA) is the primary federal authority on energy statistics and analysis. Each week, among other things, they release reports on coal production, and each month they give reports of consumption, stockpiles and make estimates of future conditions of energy markets.

I believe their current estimates on coal are significantly off, and here is why I do:

First let's take a look at published statistics for U.S. coal for 2012:

in tons Productn Elec Usage Exports Total Used Beg Inv End Inv
Jan 94,944 76,370 11,931 88,301 172,387 179,030
Feb 85,763 68,370 10,522 78,892 179,030 185,901
Mar 85,698 63,070 14,074 77,144 185,901 194,455
Apr 77,624 57,000 13,711 70,711 194,455 201,368
May 81,825 68,180 12,829 81,009 201,368 202,184
Jun 81,911 76,690 10,353 87,043 202,184 197,052
Jul 86,344 91,630 8,647 100,277 197,052 183,119
Aug 90,839 87,980 8,732 96,712 183,119 177,246
Sep 81,846 74,390 4,054 78,444 177,246 180,648
Oct 85,244 72,180 9,051

81,231

180,648 184,661
Nov 84,152 75,070 7,110 82,180 184,661 186,633
Dec 80,208 78,330 3,588 81,918 186,633 184,923
2012 1,016,398 889,260 114,602 1,003,862 172,387 184,923
Click to enlarge

All numbers, except for exports, come direct from EIA website. They give a total for 2012 exports, so it was easy to plug in these monthly amounts to make the rest of the table work (i.e., Beginning Inventory plus Production, minus Total Usage giving Ending Inventory).

Now, let's look at what they now forecast for 2013:

in tons Productn Elec Usage Exports Total Used Beg Inv End Inv
Jan 83,892 80,600 7,897 88,497 184,923 180,318
Feb 76,673 73,900 7,982 81,882 180,318 180,000
Mar 83,182 79,430 7,982 87,412 180,000 182,000
Apr 85,520 68,580 7,982 76,562 182,000 190,000
May 84,780 71,380 7,982 79,362 190,000 194,000
Jun 80,040 78,090 7,982 86,072 194,000 191,000
Jul 88,630 89,710 7,982 97,692 191,000 181,000
Aug 89,350 90,250 7,982 98,232 181,000 176,000
Sep 86,660 79,380 7,982 87,362 176,000 177,000
Oct 89,759 77,030 7,982 85,012 177,000 184,000
Nov 85,170 74,870 7,982 82,852 184,000 187,000
Dec 90,710 85,810 7,982 93,792 187,000 183,000
2013 1,022,335 949,030 95,700 1,044,730 184,923 183,000
Click to enlarge

*Total Exports is the forecast; monthly exports are not forecast, so I just divided the total by 12 to get something which is close enough for this analysis.

Now look closely at these numbers. The math in 2012 works perfectly, Beginning Inventory plus Production minus Usage equals exactly the Ending Inventory. In 2013, it does not work at all - if you do the arithmetic, ending inventories should be: 162,528 tons.

The January row is as reported in their March 22nd report. After that, Feb. and March Production and Elec usage are actual numbers as reported by the EIA. Everything else is forecast by them; again, exports are prorated by their annual forecast.

So already, by their own forecasts, they are already off by 20,472 tons of coal for the end of the year, but it is much worse than that.

Let's look at production. Thru March of 2012 actual production was: 266,405 tons, in 2013 it was 241,716, a decrease of 9.3% or 24,689 tons. For April thru December their forecast for production is 773,626 tons, an INCREASE of 3.15% over last year, after the 9.3% DECREASE in the first quarter. This cannot be correct.

In April alone this year, the first 2 weeks of the month, the production was 18,846 and 17,800 tons each, for their estimate for April (85,520) to be correct, production would have to average 21,382 tons per week for the rest of the month. Even in all of 2012, the highest production week was 20,438! If production is more like what it was for the first half of the month, April will turn out to be more like 78,527, a difference of 8,000 tons.

Let's do the math for the rest of the year. If production goes from -9.3% for the first quarter to let's say a 4% reduction from 2012 May to Dec. levels, a not unreasonable estimate, we have total production of 1,000,773 tons, another 14,569 tons off their estimate and will give us ending inventories at 147,959 on Dec. 31st, 2013.

But the biggest error in their forecasts isn't production, it's usage in the electricity generation sector. Again, actual amount of coal used for electricity for the first quarter as reported by them is: 233,930 tons, as opposed to 207,810 for the first Qtr. in 2012, an increase of 26,120 tons or 12.6%. For the rest of the year they forecast an increase of only 33,650 tons or 4.9%. If you'll recall, it was in March when the price of nat gas really started to increase, and didn't even hit $4 until the very end of the month. On Friday, April 19th, it hit $4.40 mbtu. So they have to be wildly wrong with their electric power estimates for the rest of the year. If we only increase by 12.6% (the same amount as the first quarter) for the rest of the year instead of the 4.9% they forecast, we'll use another 52,003 tons more than their forecast, leaving ending inventories at 95,936 tons.

What if gas stays at this level or goes even higher? This is very possible if you look at the weekly rig count which was 315 on April 19th as opposed to 419 rigs exactly a year ago, not to mention the fact that gas in storage is down 32% from this time last year. What if instead of using 12.6% more coal for electricity, like the last quarter, when gas was in the 3's, we used 20% more coal for electricity, not all that unreasonable if gas stays at 4.40 or goes higher. That would add another 50,000 tons to usage for the year, and inventories would be down to 46,000 tons.

If history is any guide, when inventories get down in the 150,000 ton range, prices for coal rise substantially higher than they are right now.

Conclusion: It doesn't take much of a price increase for mining companies such as Arch Coal (ACI) to be profitable. Last year Arch sold 140.8 million tons of coal for $4.2 billion and had an adjusted loss of $77 million, a scant 1.9% increase in the price of coal would have given them a small profit. If coal goes up at all, an increase of only $5 per ton could translate into an EPS of about $3. Arch closed on Friday at $4.79/share.

Inventories, will not be 46,000 tons on December 31st, but they won't be 183,000 tons either. Production will increase, and so will coal prices. One thing I know about the price of coal stocks is that they move really fast. It is not out of the question that when sentiment changes, prices could jump 20% in a week and multiply in a few short months. ACI was $4.47 a couple days ago; it was $35 a share in 2011 and $77 in 2008. The risk/reward is too much to resist.

Peabody Energy (BTU), Console Energy (NYSE:CNX) and Alpha Natural (ANR) would all benefit greatly from any increase in the price of coal. Sentiment for coal is at an all time low. These companies are priced for Armageddon. Buy when others are fearful.

Back up the truck: I'm loaded up on Arch Coal, Peabody Energy and Alpha Natural.

Disclosure: I am long ACI, BTU, ANR, WLT. 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.