Description
Amrep owns a lot of land in New Mexico. Much like California Coastal (CALC) when I wrote it up, I consider this security to be a free put on real estate values. This writeup is short for timeliness reasons.
The value of the company is in three areas. They have 95m of cash on the balance sheet, though 20m of this is likely overstated due to a temporary payables spike they discuss in the last quarter. They have a magazine distribution business which is generously worth 50m at 7x ebit. That leaves 330m of value for their land holdings in Rio Rancho, NM. Rio Rancho is a suburb of Albuquerque, and shares a lot of characteristics with the now-imploding Phoenix market: California-based investment buyers driving the market, plenty of land, and a warm desert climate.
There is not enough data to get true sales prices per acre, but it seems like around 30k/acre for undeveloped and probably twice that for developed acres. They actually don’t break land apart this way, so this is somewhat of a guess, but is reinforced by a few other data points. They further break apart land into contiguous (large lots, sold to big builders) and noncontiguous (sold to private owners or small builders); only 25% of the land is contiguous.
My basic calculation is as follows: 18,000 acres, 65k/acre, 75% gross margin (has ranged from 40-65% historically), 3% sales commission, 5m overhead/year, 40% tax rate, and a 60% discount factor (assuming they sell the land over the next 12 years, though they’ve said they’ll do it over 20). This gets me to 275m of value, as compared with the 330m ascribed to it.
In the short term, though, I think things could get pretty bad. I obtained housing starts from the local land office, and here is the data:
YEAR |
JAN |
FEB |
MAR |
APR |
MAY |
JUN |
JUL |
AUG |
SEP |
OCT |
NOV |
DEC |
TOTAL |
2006 |
322 |
178 |
337 |
242 |
170 |
191 |
126 |
113 |
81 |
|
|
|
1760 |
2005 |
193 |
347 |
195 |
335 |
288 |
315 |
254 |
317 |
236 |
218 |
146 |
240 |
3084 |
2004 |
94 |
85 |
201 |
176 |
142 |
185 |
166 |
135 |
113 |
181 |
121 |
121 |
1,720 |
2003 |
60 |
79 |
110 |
114 |
111 |
135 |
141 |
105 |
93 |
103 |
71 |
102 |
1224 |
2002 |
71 |
66 |
83 |
75 |
68 |
77 |
70 |
83 |
66 |
90 |
75 |
77 |
901 |
2001 |
62 |
67 |
102 |
78 |
114 |
83 |
64 |
92 |
68 |
72 |
58 |
56 |
916 |
2000 |
44 |
28 |
47 |
64 |
57 |
57 |
69 |
43 |
70 |
51 |
44 |
54 |
628 |
1999 |
20 |
53 |
26 |
61 |
47 |
23 |
55 |
48 |
48 |
30 |
37 |
58 |
506 |
1998 |
35 |
33 |
85 |
85 |
63 |
75 |
57 |
38 |
66 |
60 |
16 |
51 |
664 |
1997 |
48 |
70 |
69 |
35 |
87 |
79 |
18 |
63 |
23 |
43 |
49 |
91 |
675 |
1996 |
34 |
81 |
58 |
48 |
65 |
64 |
91 |
110 |
40 |
91 |
38 |
58 |
778 |
1995 |
56 |
60 |
105 |
77 |
117 |
63 |
56 |
67 |
55 |
64 |
50 |
89 |
859 |
1994 |
63 |
72 |
69 |
64 |
41 |
52 |
41 |
65 |
67 |
91 |
56 |
91 |
772 |
1993 |
29 |
73 |
105 |
91 |
90 |
115 |
80 |
100 |
48 |
111 |
70 |
91 |
1,003 |
1992 |
22 |
43 |
59 |
54 |
48 |
55 |
69 |
81 |
72 |
24 |
52 |
41 |
620 |
1991 |
35 |
31 |
39 |
39 |
31 |
44 |
45 |
43 |
48 |
49 |
95 |
75 |
574 |
1990 |
29 |
20 |
14 |
25 |
53 |
39 |
30 |
76 |
29 |
35 |
38 |
28 |
416 |
1989 |
58 |
65 |
105 |
64 |
61 |
59 |
51 |
53 |
37 |
49 |
8 |
46 |
656 |
1988 |
38 |
75 |
100 |
70 |
99 |
127 |
54 |
58 |
103 |
55 |
45 |
39 |
863 |
1987 |
102 |
106 |
95 |
78 |
36 |
48 |
43 |
39 |
84 |
81 |
43 |
45 |
800 |
1986 |
73 |
85 |
89 |
118 |
74 |
129 |
76 |
95 |
109 |
74 |
72 |
32 |
1,026 |
1985 |
91 |
108 |
67 |
73 |
182 |
105 |
93 |
103 |
81 |
128 |
37 |
59 |
1,127 |
TOTAL |
|
|
|
|
|
|
|
|
|
|
|
|
15,008 |
This table probably won’t format correctly, but starts were 500-1,000 a year through 2002, when they spiked to 1,200 for 2003, 1,700 for 2004, and 3,100 for 2005. in 2006, the monthly progression since january has been 322 178 337 242 170 191 126 113 81. This last number is for September; October comes out next week, though the initial read from the land office is “lower than September”.
In addition, here are some quotes from local brokers from a few weeks ago. They’re basically what you would expect in this market environment.
"Softening in that we now have a buyer's market (in excess of six months house inventory."
"Because of the large inventory of resale homes and the higher interest rate, builders have slowed up their new projects. Many are on the drawing board and most in Rio Rancho, but the builders are waiting until Spring. We do have a winter here, albeit mild, so it is best to wait. Many of the homes for sale in RR are being sold by investors so that is why you see a lot on the market."
"As talked about indirectly above, the market is slowing and so more “For Sale” signs are coming up. Also, many sellers who are really needing to move are starting to make big price decreases."
Finally, the main employer in Rio Rancho is Intel, which has announced a major headcount reduction initiative. According to Intel, Rio Rancho is safe until mid 2007, but perhaps could see cuts then. Since Intel is building a new fab in Arizona, they won’t be hiring in Rio Rancho for a while.
AXR is an IBD top momentum stock. I believe, because of the strong decline in home starts, that no builders were buying lots (either outright or through option exercises) in August, September, or October, and thus earnings could be under $.50 (versus $2.38 in the last quarter), which would boot the stock of the IBD and break its momentum.
The risks are that the company is controlled by a few families, who own 75% and could always take the company private. Offsetting this, I would argue that they should report a bad quarter first and then buy the company, not vice versa. Also, they've owned this stock for over a decade, so they could decide to monetize some of it as well.
Catalyst
significant decline in earnings
broken momentum