Unlock the Editor’s Digest for free
Roula Khalaf, Editor of the FT, selects her favourite stories in this weekly newsletter.
The writer is a bond portfolio manager at Barksdale Investment Management and editor of ‘The Credit Investor’s Handbook’
When Elon Musk’s SpaceX brought its inaugural bond deal to market last week, credit investors took the company’s presentation seriously but not literally.
After obtaining investment-grade credit ratings from all three of the rating agencies, it was priced at the cheap end of this part of the debt universe, but at a meaningfully better level than Oracle, the tech company known for its level of cash burn among investors.
Investors appeared not to take some of the more novel parts of the offering memorandum, such as an “Artist Visualisation of Life on Mars” at face value. Instead, the creditor-friendly profile of Starlink, the satellite business that is SpaceX’s cash cow, drove credit ratings and bond pricing.
Creditors also took comfort from the lip service of chief financial officer Bret Johnsen to balance-sheet discipline. SpaceX indicated a target for its debt of 2-3 times earnings before interest, tax, depreciation and amortisation (versus about 5 times today) as it grows into its capital structure.
Such an approach is not unprecedented in the bond market; for example, Netflix embraced a similar narrative in its communications with creditors as it crawled its way out of a cash-burn business model. Netflix, however, started its life as a bond issuer firmly in junk ratings territory and did not achieve a fully investment-grade rating until more than a decade after its inaugural bond issue.
Although there was little clear guidance on capital expenditure from SpaceX, or when it might break even on a free cash flow basis, its commitment to a minimum liquidity buffer of $25bn and a reference to its ability to adjust capex as needed soothed bondholders. As Moody’s put it, sternly: “The rating incorporates our expectation that SpaceX would adjust its investment pace in the event of material setbacks.”
You might reasonably ask how shareholders and creditors draw quite different conclusions from what are very similar offering memorandums. Unlike its bond spreads, SpaceX’s stock price implies a stratospheric valuation as a multiple of revenue or ebitda.
Creditors are pricing the certainty of cash flow coverage by Starlink, while shareholders are focused on the virtually infinite growth potential of SpaceX’s xAI. The AI business has negative ebitda and managed to burn roughly $20bn in cash over the past 12 months.
Both equity and bond investors are taking leaps of faith that might one day be at odds with each other: that SpaceX will turn off the capex tap as needed, and that it will throw money at pie-in-the-sky plans like data centres in space, regardless of cost, to preserve its first-mover advantage.
In an ideal world (or, a less accommodative credit market), creditors would lend directly to the Starlink business, the group’s primary profit driver which generated more than $7bn of ebitda in 2025, and threw off close to $3bn in cash flow, excluding interest payments. Moody’s acknowledged the importance of Starlink by using its “telecom services provider ratings methodology” to underpin its Baa1 rating.
Shareholders, less interested in the boring merits of Starlink, appear to be taking parts of the offering memo more literally; perhaps they believe, for example, SpaceX’s claim to have “identified the largest actionable total addressable market in human history” of $28.5tn. Is it a coincidence that this corresponds roughly to US GDP? Had creditors used this “forecast” they might have priced SpaceX at a negative spread to US Treasury bonds.
The variety of outcomes possible for SpaceX has value that accrues almost entirely to shareholders. If it captures even a third of the total addressable market outlined by the company, its share price will be justified. In that scenario, the spread on the yield of SpaceX bonds over benchmarks would likely tighten, but the upside is capped by basic bond maths. Shares can rise to the stratosphere but yields can only compress so much. And creditors who hold to maturity will only ever get par back.
On the flip side, creditors who lived through the period when Musk’s Tesla almost went bankrupt over 2017-19 and its bonds traded as low as nearly 80 cents in the dollar, know all too well that his eccentric genius is at best a double-edged sword for creditors. Taking SpaceX’s capital structure and liquidity guidance seriously is about as far as creditors should be willing to go at this point.
They have appropriately left the literal reading — and some might say magical thinking — of the bond offering to the stock market. The creditors must be prepared to remind SpaceX of the seriousness with which they take its balance sheet commitments.
Barksdale may hold interests in securities mentioned in this article
