Accelerant

Accelerant's marketplace proposition should be assessed on two separate dimensions: who issues the policy and who ultimately retains the underwriting exposure. The company reported that third-party direct writers accounted for 47% of second-quarter 2026 exchange premium, up from 27% a year earlier. Its own retained share also increased, from 6% to 13%. Those company-reported measures describe different parts of the structure. More outside issuing capacity does not, by itself, demonstrate that Accelerant is taking less underwriting risk.

For an MGA, that distinction has practical consequences. Joining an exchange does not eliminate the need to identify the contracting carrier, delegated authority, reinsurance dependencies and renewal terms. Our assessment is that the platform's strongest potential advantage is a reduction in the effort required to connect a specialist portfolio with suitable capital. That advantage needs to survive a change in appetite. A useful diligence exercise would trace what happens to one program if an issuing insurer or risk-capital participant withdraws, including who controls the data and which approvals a replacement would require.

The August 2026 results release also describes ARC, a tool intended to recognize and structure incoming data. That is a stated capability, not evidence of lower losses or fewer underwriting errors. In our view, the important controls are traceability to the original submission, a record of corrections and a clear boundary between extracted information and underwriting judgment. Comparable data can make portfolios easier to evaluate, but forcing dissimilar risks into the same fields can also hide differences that matter. The test is whether the information remains useful when claims develop, not simply whether it enters the system faster.

Scale creates another analytical challenge: exchange-wide performance can conceal variation between members and lines. A capacity provider needs to understand which cohorts are improving, which are deteriorating and whether apparent diversification reflects genuinely different exposures. For an MGA, the corresponding question is how its own experience affects pricing, fees and available support. These are diligence questions rather than findings of weakness. Growth and platform earnings alone cannot establish that every participating portfolio has sustainable underwriting economics.

On August 13, Accelerant announced an agreement to be acquired by Thoma Bravo at an enterprise value above $4 billion. The announcement anticipated completion in the first half of 2027, subject to closing conditions and approvals; it did not announce a completed acquisition. A software-focused owner's proposed involvement is relevant to the technology strategy, but it does not establish software-like risk characteristics for an insurance business. Our view is that the durable test remains the same under either ownership structure: reliable information, clear accountability and capital relationships that continue to work when results disappoint.

Accelerant’s second-quarter 2026 SEC segment disclosure separates Exchange Services, MGA Operations and Underwriting. Exchange Services charges participating insurers for sourcing, managing and monitoring business; MGA Operations includes Mission and Owned Members. Transactions between consolidated affiliates are eliminated when the group reports its consolidated results. This matters because adding the individual businesses’ revenues would overstate the external revenue of the group. The segment presentation is a view of operating components, not three independent revenue streams that can simply be summed.

The filing reports second-quarter adjusted EBITDA of $74.0 million for Exchange Services, $30.3 million for MGA Operations and $1.9 million for Underwriting. Corporate and Other and consolidation adjustments bring the consolidated figure to $93.1 million. These are company-reported non-GAAP measures, not measures of underwriting profit or available claims-paying capital. The company also changed its adjusted EBITDA definition in the first quarter of 2026 to exclude realized and unrealized investment gains and losses, with prior-period segment information conformed. Comparisons should use that consistent presentation.

MGA Index analysis: for a prospective member, the useful diligence exercise is to trace a single program through this structure. Identify which entity receives the service fee, which entity earns the underwriting commission and which balance sheet retains losses after reinsurance. Then test what changes if premium falls, claims deteriorate or an external insurer withdraws. A growing fee business and a stable capacity relationship are related, but they are not the same proposition. Conversely, a small underwriting-segment earnings contribution does not by itself show weak underwriting: retention, reinsurance, expense allocation and the accounting period can all affect that contribution. The point is to establish the mechanism before interpreting the headline result.

A second distinction concerns growth. In its August 13 results release, Accelerant reported 314 members at June 30, compared with 248 a year earlier, and net revenue retention of 111%, compared with 151%. Those figures should not be described as policyholder retention or as the percentage of members that renewed. Our analytical question is how much expansion comes from existing relationships versus newly added ones, and what happens to economics as those cohorts mature. The reported figures alone do not establish why the retention measure changed; an explanation requires the company’s metric definition and a fuller cohort bridge.