Berkshire Reports Q1 (Bore Port) May 17, 1999

<THE BORING PORTFOLIO>
Berkshire Reports Q1

by Dale Wettlaufer (TMF Ralegh)

ALEXANDRIA, VA (May 17, 1999) -- Berkshire Hathaway (NYSE: BRK.A) lost $2000 to $72,300 today after releasing its first quarter results on Friday afternoon after the bell. The company released its interim shareholders' report on its website over the weekend. As in all quarters, GAAP net income doesn't mean that much from an analytical viewpoint, since any company with unrealized gains on the balance sheet and huge insurance operations can manufacture all sorts of earnings in the short term.

Berkshire could, for instance, realize tens of billions of dollars in gains tomorrow and report it for the second quarter, but that wouldn't mean too much. Therefore, you have to look at the operating income. For the purposes of looking at the consolidated company, I've reconstructed last year's first quarter, but the pro-forma results include only GenRe for Q1 '98.

Q1 '98:

Net income of ($722 million + $272 million) less realized investment gains of ($723 million + $37 million) and add back taxes at 37%. That's $994 - $760 + $281.2 = $515.2 million. Now add back goodwill amortization of $23 million plus $7 million. That's net income before goodwill amortization and realized gains of $535.2 million.

Another way to work through the numbers, which would give us Berkshire, GenRe, and Executive Jet for 1998, is this: Net income of $879 million less realized investment gains of $723 million + $37 million and add back taxes at 37%. That's $879 million minus $760 + $281.2. Add back reported goodwill amortization from last year of $23 million and $7 million, plus incremental goodwill amortization of $90 million from the Gen Re purchase, plus another estimated $3 million in Executive Jet goodwill amortization. That gets you to $523.2 million in net income before realized gains and goodwill amortization for Q1 '98. Per share, that's $344.63. The discrepancy would be due to Executive Jet's losses as well as seven days of Dairy Queen's Q1 '98 results, as that acquisition closed on January 7, 1998.

Incidentally, as a way to check revenues on Executive Jet Aviation (EJA), we can see that the pro-forma results for Q1 '98 given in Berkshire's quarterly report indicate no intra-company eliminations. Pro-forma earned premiums for Q1 '98 are indicated to be $2,841 million and actual earned premiums at Berkshire and Gen Re for that quarter were $1,367 million + $1,474 million = $2,841 million. The difference between pro-forma Q1 '98 revenues of $5,345 million and reported Q1 98 revenues of $3,325 million + $1,925 million = $5,250 million is $95 million. Reduce that by what I would estimate as $6 million in revenues for the first seven days of January at Dairy Queen and you get $89 million in revenues at EJA for Q1 '98.

With Flight Services revenues of $105 million last year and $431 million this year, I assume both FlightSafety and EJA had a huge quarter. I don't think it's unreasonable to assume FlightSafety could have grown 35% year-over-year. If that were the case, then EJA grew 225% year-over-year, with revenues increasing from $89 million to $289.2 million. Of course, FlightSafety could have had an even bigger quarter, but even if you tweak the assumptions a bit, the NetJets program at EJA looks to be going great guns.

On operating income for Q1 '99, we have net income of $541 million less realized gains of $403 million and add back taxes at 37%. That's $541 minus ($403 + $149.11) = $287.11. Add back goodwill amortization (I assume in both cases it's not deductible) of $118, and you get net income of $405.11 million. Per share, that's $266.71. Per-share operating earnings dropped 22.6%, which isn't inconsistent with what you would expect with global reinsurance being as soft as it is.

According to Berkshire's first quarter 10-Q, GenRe reported a pre-tax underwriting loss of $136 million this quarter and a slight underwriting gain of $17 million for Q1 1998, so the swing is $151 million year-over-year. In addition, GEICO is pouring gas on the customer acquisition fire:

"Premiums earned by GEICO during the first quarter of 1999 exceeded amounts earned during the first quarter of 1998 by 17.5%, reflecting continued growth of voluntary auto policies-in-force, partially offset by the effects of premium rate reductions taken over the past two years. In-force policy growth of 22.5% during the past year is attributed to accelerated levels of advertising and competitive premium rates. It is expected that the numbers of voluntary policies in-force will continue to grow at a substantial rate over the remainder of 1999."

Just as at Amazon.com, you take the earning hit up front if the customer is worth much more in the long run. Although GEICO shows no underwriting income, the additions to deferred premiums and other float items at no cost adds tremendous value. The major growth drivers at Berkshire are in-place and are very encouraging. While it's not fun to sit around and watch global reinsurance conditions depress these other elements of progress, I personally believe that GenRe was acquired at a very good price, that the people at GenRe are motivated to do well with the things they can control and not do stupid things in reaction to conditions they can't control, and that the merger with GenRe was a value-adding move.

Finally, on earnings, there's another way of looking at how the company did this quarter. As I mentioned last week, check out lecture number two from the following series of lectures Ben Graham gave at the New York Institute of Finance more than 50 years ago: http://www.wiley.com/products/subject/finance/bgraham/. This is what Ben Graham called the "comparative balance sheet method." If you want to reference this in older editions of Security Analysis, it would be under "balance sheet: as a check on earnings."

What I want to look at is the change in shareholders' equity for the quarter, which eliminates some of the GAAP earnings problems. This automatically wipes out the problem, for instance, of the quarter's realized gains. It also recognizes cash paid for income taxes, which was well in excess of reported income tax expense for the quarter. It also recognizes some items that you would call "comprehensive income," which over the years have been very real in measuring Berkshire's economic progress, such as increases in the value of securities.

It does have the problem, however, of being dependent upon GAAP balance sheet estimations, which in the hands of shadier elements is just as dangerous as looking at overoptimistic estimations of likely insured losses and the like. In other words, it's nowhere near perfect, and the only time I've ever looked at a company in this manner was last week when I wanted to get a better grip on Conseco's financials. (No, don't worry, I'm not interested. Just a masochistic exercise in financial statement analysis). So here's a quick pass at it.

Quarter-to-quarter change in shareholders' equity was $481 million. Deduct from that $17 million that came into paid-in-capital. (I would guess through the exercise of GenRe options, but I'm not sure there. In any case, throw that out.) So, the change is $464 million. Here's where you see the realized investment gain has no bearing on shareholders' equity. The increase to retained earnings effected by that is offset by a decrease in accumulated other comprehensive income. On an assets-liabilities basis, this holds true, because the assets change form, going into cash with a little commission haircut while the decrease in the deferred tax account is offset by a decrease in cash to pay the taxes.

So the next item you want to deal with is the reduction to retained earnings that arises from goodwill amortization. Since goodwill is pretty much the evidence of deployed equity, and since the $118 million reduction in the asset is meaningless in that context, that gets added back to the change in owners' equity. So we have $464 million plus $118 million goodwill amortization = $582 million. As for appropriation of surplus for insurance reserves, that will be reflected in the change in shareholders' equity. Again, I'm making no adjustment for that because I trust the company's judgment on that. That takes care of any major changes in the balance sheet period-to-period, so the $582 million is the estimate of economic income yielded by my working through this method.

Since realized investment gains are wiped out of this calculation, it's a total coincidence that GAAP earnings are in the neighborhood of this figure.

By the way, since I am a Fool, I'm sure I've screwed up somewhere above. I'm open to corrections and comments and any other input you might have on the Boring message board (some people would say that's an apt name this evening). Hope to see you there. Best of luck to the Buffalo Sabres.

• Discuss Boring Investing on the Boring Port message board.

Call Your Boss a Fool.

Stock Change   Bid
APCC    ---    32.38
BRKb  -51      2347.00
CSL   -  1/16  46.19
CSCO  +1       116.44
GTW   +  1/2   66.50

                  Day     Month   Year  History
        BORING   -0.21%  -1.60%   3.23%  38.61%
        S&P:     +0.13%   0.32%   9.29% 122.83%
        NASDAQ:  +1.34%   0.75%  16.84% 146.10%

    Rec'd   #  Security     In At       Now    Change
  6/26/96  225 Cisco Syst    23.96    116.44   386.06%
  8/13/96  200 Carlisle C    26.32     46.19    75.45%
  4/20/99  230 American P    28.95     32.38    11.82%
 12/31/98    8 Berkshire   2244.00   2347.00     4.59%
   2/9/99  100 Gateway 20    72.38     66.50    -8.12%


    Rec'd   #  Security     In At     Value    Change
  6/26/96  225 Cisco Syst  5389.99  26198.44 $20808.45
  8/13/96  200 Carlisle C  5264.99   9237.50  $3972.51
 12/31/98    8 Berkshire  17952.00  18776.00   $824.00
  4/20/99  230 American P  6659.25   7446.25   $787.00
   2/9/99  100 Gateway 20  7237.50   6650.00  -$587.50


                             CASH    $999.27
                            TOTAL  $69307.46