Investors loved Boeing’s fourth-quarter 2014 financials. Operating cash flow hit $8.86 billion, a figure projected to exceed $9 billion this year. Boeing says it will keep returning 80% of free cash to its investors. The company’s stock jumped 5.4% right after the earnings announcement.

But scratch a little deeper and there’s a serious problem: Boeing’s all-important commercial jetliner unit would be far less profitable without deferred production cost accounting. Worse, there are few signs that this situation will change soon. The company continues to lose money on each 787 Dreamliner it builds. These recurring production losses (separate from development costs) increased by $900 million in the last quarter, reaching $26.1 billion. If those losses were not carried forward and amortized over a large and growing block of future production aircraft, deferred costs from 2014 would be enough to wipe out much of the profits earned from the other 609 jetliners that Boeing delivered last year.

To understand the extent of these losses, consider the impact on future production. Right now, Boeing is using a 1,300-aircraft accounting block for the 787. Since the company says it will start earning money on each one later this year, let’s assume 1,000 future 787s will bear this burden (247 had been delivered as of Feb. 28). If we assume deferred costs top out at $28 billion, that means a $28 million tax on each of these 1,000 planes.

According to several valuation sources, Boeing’s realized price for an average 787 is about $120 million. Removing that $28 million deferred-costs levy leaves $92 million per aircraft. Since the company will want to maintain a respectable 10% profit margin, that accounts for another $12 million off each plane. Boeing, therefore, will need to reduce total manufacturing costs (in its plants, and from its many suppliers) down to $80 million per aircraft in 2015 dollars.

This sounds difficult, to say the least. Worst of all, the bill will be coming due just as commercial deliveries are reaching plateaus and possibly declining, and the three most profitable defense programs (C-17, EA-18G, F-15 and F/A-18E/F) are winding down.

This grim 787 program profitability outlook explains Boeing’s aggressive actions with its workforce and supplier base. With its heavy-handed approach to labor contracts, linked to threats to move the 777X line and its efforts to renegotiate supplier contracts under its Partnering for Success program, Boeing has done everything it can to cut costs.

Boeing, confident these efforts will produce results, continues to offer an optimistic outlook on 787 costs. In its fourth-quarter 2014 conference call, the company said it expected “the 787 to be cash-positive during 2015 and we still anticipate deferred production to decline shortly after we’ve achieved the 12 per month production rate in 2016.”

There are three reasons to be skeptical of this forecast:

•The first is Boeing’s track record. Originally, deferred costs were projected to top out at $20 billion. The company then moved its guidance to a peak of $25 billion, but 2014 figures broke through that barrier, and 2015 will see at least another $2 billion added.

•The second is that the trend line (see graph) appears to be stuck at just above $30 million in losses per plane. Design changes associated with 787-9 variant, which is just ramping up to high-rate production, will likely disrupt the learning curve.

•But the biggest reason to be skeptical is that there is nothing new in Boeing’s approach to the problem. Management believes if it continues to squeeze suppliers and labor, the problem will be solved. Again, the track record here is not great.

Most of the manufacturing world tells a very different story. Whether it’s with cars, aircraft or turbines, productivity improvements often come from the shop floor. That means convincing the people who build things to identify ways to reduce scrap, improve work flow and eliminate defects.

To promote the kind of process improvements that happen in the factory, a workforce needs incentives such as profit-sharing or other compensation. At the very least, machinists and engineers need to believe their work is valued. Taking away pensions at a time of record sales is simply a bad way to motivate workers to go the extra mile. Boeing right now embodies a strange combination of very good and very bad. The company’s commercial unit is enjoying record revenue in a very strong market. It holds the top spot in a duopoly with high barriers to entry, and has the world’s best jetliner product line.

But whoever inherits the CEO mantle from Jim McNerney, who is expected to retire in the next year or two, will also inherit a somewhat toxic legacy: awful labor relations; and $28 billion, and quite possibly more, in delayed financial pain.

Richard Aboulafia is vice president of analysis at Teal Group. He is based in Washington.