Calpine's

(CPN)

capital needs are going up as its earnings go down.

A financial filing Thursday, released after a hefty earnings warning on Wednesday, shows the power producer is thinking of raising another $900 million in the markets, even though it already has tapped various sources for more than $5 billion in the last three months.

But a close look at Calpine's cash requirements suggests it still needs every penny it can lay its hands on. Needless to say, that sends a worrisome signal to investors leery about the power industry in the wake of the collapse of

Enron

.

Powering Down

The San Jose, Calif.-based company took its forecast for 2002 earnings down by a third from $2.45 to $2.60 per share to $1.70. And it also said it was lowering its 2001 forecast to $1.95, from a range of $2 to $2.05. The reductions were widely expected and are chiefly attributable to a drop in power market profit margins and a decline in electricity demand.

Power profits are of critical importance to Calpine, which is in the process of racking up some $15 billion of debt to build a massive collection of cutting-edge power plants. If those plants don't produce strong returns in 2002, the company will struggle to pay its debt costs and to carry out the capital expenditures needed to achieve the company's newly lowered target of having 26,300 megawatts of generation capacity in operation by 2003. It currently has 11,100 megawatts in operation.

Despite a lengthy conference call Wednesday in which Calpine executives laid out expected cash flows, the market wasn't comforted. The stock dropped a dollar to $13.95 Thursday, leaving it a goading 75% below its 52-week high.

The table below shows Calpine's expected uses and sources of cash in 2002. The debt service costs, which the company declined to break out, are embedded in the $1.2 billion operating cash projection. Clearly, the company doesn't have much margin for error, which may be why it's thinking of coming back to the market.

"It's a little too snug," Calpine investor relations chief Rick Barazza says of his company's current funding situation. "We'll be coming back to the market." But Barazza says Calpine is eager to win back its investment grade rating.

Moody's rated Calpine debt at investment grade from October to December 2001, while Standard & Poor's never had the credit in that category. That desire for a better rating could lead Calpine to issue equity, Barazza says. The Thursday filing provides for a range of financing vehicles, including common and preferred stock and warrants, which are instruments that give investors the right to buy shares.

The market's sour reaction Thursday could betray a lack of confidence in the cash projections. For instance, the $1.2 billion operating cash flow may not materialize. This number excludes debt service costs, but the company won't give those out. It does offer a figure for earnings before interest, taxes, deprecation and amortization, or EBITDA. That's likely to be $2 billion in 2002, says Barazza. The difference between EBITDA and operating cash -- $800 million -- is the total number for cash taxes plus cash interest. (Calpine doesn't run a big chunk of its interest payments through the income statement, so it's hard to get a solid cash number for that from its financials.)

Abacus

The numbers don't add up. Calpine's construction debt totals $3.5 billion, and bears an interest rate just under 5%, while its $7 billion to $8 billion of other debt costs about 8.5%, Barazza says. That would put total interest at around $800 million, or exactly the difference between EBITDA and operating cash flow. There seems to be no room left for the tax expense. And Calpine's debt will be a lot higher than $11.5 billion in 2002, because the company hasn't drawn down many of its credit lines.

Calpine also cut the amount of plant it still intends to construct from around 17,000 megawatts in December to 15,200 megawatts Wednesday, a drop of 1,800 megawatts. Barazza says the drop is explained by plans to re-evaluate three projects that Calpine is "just breaking ground on." At a cost of $550,000 per megawatt of capacity, that would save the company $1 billion. Barazza denies that the cut was provoked by cash constraints.

Of course, the main driver of operating cash flows for Calpine is what it earns from its power sales. The Calpine boosters, including 90% of all Wall Street analysts who follow the company, used to say that because Calpine has arranged long-term contracts for most of its sales, there was a great deal of reliability in its earnings forecasts.

But that theory took a body blow Wednesday, when the company slashed guidance. The 2001 cut was the most eye-opening. Calpine had hedged 90% of its sales last year, and it still had to slash guidance. The hedged portion is 65% for 2002. The margin on the nonhedged sales will be much lower, due to a drop in natural gas and power prices over the past year.

Spark Plugs

Calpine did provide a forecast margin, or spark spread, for unhedged sales Wednesday, but declined to do the same for the much larger proportion of sales that are hedged. Barazza also refused Thursday. "How many other companies do that?" he asks. And he added that margins could improve as natural gas prices, languishing at low levels, rise. (Calpine benefits if gas and power prices move up together.)

The spark spread for the hedged sales is crucial to Calpine's fate in 2002. Part of the reluctance to give out the spread is that it may reveal the fat profits the company is making from large long-term contracts forged with California in the midst of the energy crisis. Calpine is renegotiating contracts with the state. But these talks may result in Calpine losing hedged revenue as the state demands price concessions. State auditor Elaine Howle was quoted in the

Sacramento Bee

as saying her consultants thought "a couple of the Calpine contracts were two of the worst contracts the state has."

It's not hard to see why the auditor's office feels this way. A contract obtained from the California Department of Water Resources shows Calpine demanding a $80 million to $90 million annual fee just for the right to have the capacity available at a peaker plant, which is designed to provide juice to a network at times of heavy use. Though peaker contracts are more expensive, this one looks steep. And the state also has to pay for the power delivered by the plant. Barazza denies that it's an excessive fee. "That is a number that is out there in other peaking contracts," he explains. A call to Howle's office wasn't returned.

Investors could be in for a shock.

Know any companies that the market may be misvaluing? Detox would like to hear about them. Please send all feedback to

peavis@thestreet.com.

In keeping with TSC's editorial policy, Peter Eavis doesn't own or short individual stocks. He also doesn't invest in hedge funds or other private investment partnerships.