While commercial cybersecurity stocks rallied to five-year-high multiples this year, the government contractors that run cyber work for federal agencies went the other way. As of October 5, 2026, Booz Allen Hamilton, Parsons, and Leidos trade at the 3rd, 1st, and 5th percentile of their own five-year EV/Sales range. CACI, doing broadly similar work, sits at the 98th. Most of that split has a plain explanation, and one name fits it less well.
If you only read one thing
The short version: Booz Allen, Parsons, and Leidos trade near five-year lows relative to sales, while CACI trades near a five-year high. The federal budget is moving money toward military cyber and away from civilian cyber. Booz Allen and Parsons are shrinking 7%, which explains much of their discount. Leidos is still growing slightly and is the hardest to explain.
The numbers
| Ticker | Company | EV/Sales | 5y median | EV/Sales 5y %ile | Revenue growth (TTM) |
|---|---|---|---|---|---|
| CACI | CACI International | 2.0x | 1.42 | 97.9 | 11% |
| SAIC | Science Applications Intl. | 1.1x | 1.04 | 53.3 | −1% |
| LDOS | Leidos | 1.1x | 1.37 | 4.8 | ~4% |
| BAH | Booz Allen Hamilton | 1.1x | 1.70 | 2.9 | −7% |
| PSN | Parsons | 0.9x | 1.36 | 0.6 | −7% |
These multiples look tiny next to the software names (CrowdStrike trades at 51 times sales). That is normal: government services firms bill for people's time on thin margins, so a dollar of their revenue is worth much less than a dollar of software subscription revenue. Comparing them with security software tells you nothing. Comparing each with its own history does, which is what a percentile is for. Our guide on what a valuation percentile is and how we compute it explains the method.
The takeaway: Government contractors always trade at low multiples because their margins are thin. The useful comparison is each company against its own past, and on that basis three of the five sit near five-year lows.
The budget is split
Federal cyber money is shifting between agencies. The Pentagon requested $20.5 billion for cyberspace activities in fiscal 2027, and Cyber Command's leadership described operations ramping up to meet a growing threat. On the civilian side, federal cyber spending dips from $12.46 billion to $12.23 billion, including a $707 million cut to CISA, the agency that protects civilian federal networks.
So "government cyber" isn't one trade. A contractor weighted toward defense and intelligence work sits on the growing side of the budget. One weighted toward civilian agencies sits on the shrinking side. Each company's annual report breaks revenue down by customer, and that mix is the first thing to check.
The takeaway: The defense side of federal cyber spending is growing while the civilian side is being cut, including a $707 million reduction at CISA. Which side a contractor depends on matters more than the word "cyber."
Revenue explains much of the gap
The ranking lines up closely with growth:
- CACI grew revenue 11% over the trailing year and trades at 1.95 times sales, near the top of its five-year range. The market is paying up for the growth.
- SAIC is flat (down 1%) and trades right at its five-year median.
- Booz Allen and Parsons both shrank 7% and trade near their five-year lows. A shrinking services business usually gets a shrinking multiple.
- Leidos is the exception. Revenue grew about 4% and our data shows growth picking up slightly, yet it trades at the 5th percentile on EV/Sales. Its P/E of 11.0 sits at the 3rd percentile of its own range too.
The takeaway: Growth explains most of the ranking: CACI is growing and expensive, Booz Allen and Parsons are shrinking and cheap. Leidos is growing slightly but priced near its lows, which makes it the name most worth a closer look.
Why these aren't on the cybersecurity screen
Security is a minority of revenue for all five companies, so they behave more like government services stocks than like security vendors. That is why our cybersecurity valuation screen covers the security vendors only, while the government contractors sit in their own group on the AI cyber defense theme page. If you want the full picture across both groups, our ranking of AI cybersecurity stocks by valuation percentile puts all of them in one table.
What to check before reading too much into it
- Customer mix. How much revenue comes from defense and intelligence versus civilian agencies.
- Backlog and book-to-bill. A shrinking business with a growing backlog is different from one with a shrinking backlog.
- Final appropriations. The budget figures above are requests. Congress decides the final numbers.
- Contract awards. Large recompetes can move a contractor's revenue by several points in either direction.
The takeaway: A near-record-low multiple on a government contractor is a reason to read the customer mix and backlog, not a conclusion. The budget favors defense cyber over civilian cyber, and each company's filings show which side it is on.
Data: SEC EDGAR XBRL filings and market prices, computed by StockResearch as of October 5, 2026. Budget figures come from the linked sources and reflect requests, not final appropriations. Jake is the founder of StockResearch.app, where he writes data-first research on valuation dislocations. Nothing here is investment advice or a recommendation to buy or sell any security. Do your own research.