The Government Accountability Office (GAO)—Congress’s investigative watchdog—usually gets at least grudging cooperation from executive branch departments and agencies when conducting its audits. Yesterday, GAO released a terse two-page “review” of the Office of Personnel Management’s (OPM) proposed reductions in force (RIF) and reorganization plans and activities that revealed how OPM reacted to GAO’s inquiries:

To supplement GAO’s analysis of publicly available data, GAO requested documentation from OPM regarding changes made to offices or programs since 2025, the rationale behind the closure or consolidation of offices, and processes for strategic workforce planning, among other information. Other than comments on a preliminary draft of this report, OPM did not provide any requested documents or information, nor did it agree to meet with GAO or respond to written questions. As a result, in this report, GAO is unable to provide complete information on what changes OPM made, its rationale, the expected costs and benefits, and any effect on OPM’s ability to fulfil its mission. 

Read the rest of this post →

“Democratic socialists” are a growing movement on the political left. And many of them are also prison abolitionists. But there is a deep contradiction between their socialism and their abolitionism. As the history of actual socialist states shows, serious attempts to implement socialism lead to vastly greater incarceration (and other punishment), not less. There are several ways to try to escape this contradiction. But none work. The way to reduce incarceration is to reduce the range of activities we make illegal, not increase it.

What is Socialism?

“Socialism” here means government control of all or most of the economy, not simply an increase in welfare spending or a marginal increase in certain discrete types of regulation. That’s what actual leading socialists say they want. For example, the platform of the Democratic Socialists of America (DSA), America’s most prominent socialist organization, states that it seeks to establish a system in which “[f]ood, education, energy, medicine, and transportation aren’t for-profit businesses; they are common goods.” That requires a complete government takeover of these massive parts of the economy. 

Read the rest of this post →

For years, the Jones Act has effectively cut off Puerto Rico’s access to mainland propane. The reason is simple: there are no Jones Act-compliant oceangoing LPG (liquefied petroleum gas) tankers capable of transporting it. The only possible uncertainty was whether Puerto Rico wanted American propane in the first place. Thanks to the March 2026 Jones Act waiver, that’s now settled.

Since the waiver took effect, Puerto Rico has imported roughly 809,000 barrels of mainland propane. That’s 4,792 percent more—nearly 49 times—the average annual volume shipped from the mainland during 2004–2025. Put another way, it’s more than twice the amount of mainland propane shipped to Puerto Rico during the previous twenty-two years combined.

Read the rest of this post →

New Fed Chairman Kevin Warsh testified before the House Financial Services Committee and the Senate Banking Committee last week, delivering the Fed’s semiannual Monetary Policy Report. Both sessions covered important topics: a favorable June inflation report that Warsh declined to treat as “mission accomplished,” the Fed’s retreat from forward guidance, and pointed exchanges over the Fed’s independence from the White House. But an outsized share of the questioning, especially in the Senate, went to artificial intelligence (AI), leaving some key questions about the state of monetary policy unasked.

Obviously, the Fed has no direct control over how AI evolves. But senators pressed Warsh on whether the AI investment boom is inflationary and on the financial-stability risks posed by the newest AI models. His answer was that AI-driven price increases need not be inflationary, because supply should eventually catch up, and he described the internal Fed debate over AI spending as “one of the good family fights.”

Read the rest of this post →

Republicans have one remaining reconciliation bill before the midterms. They should use it to prove they’re serious about reining in excess federal spending and fixing the debt, rather than embracing a “go small and go home to campaign” approach (as coined by the Wall Street Journal editorial board).

That means offsetting any new spending with meaningful reforms that reduce waste, fraud, and abuse in federal welfare programs, reining in deficits by slowing the growth of entitlement spending, and easing the inflationary pressures and rising interest costs that excessive federal borrowing imposes on American families. At a minimum, Congress should deliver the $600 billion in savings House Republicans pledged in last year’s reconciliation framework but ultimately failed to achieve.

Read the rest of this post →

In a strange chapter of his new book, Communion, Vice President JD Vance laments how the “technocratic cost-benefit analysis” of economics can “crowd out your sense of right and wrong.”

Vance recalls how, during a law school seminar he attended, economic reasoning was used to debate whether chattel slavery had damaged the Southern economy or instead provided output fueling the Northern states’ growth.

To Vance, that discussion exemplified an economic discipline so absorbed by efficiency and output considerations that it could only examine an obvious moral abomination as though the decisive question were whether it raised GDP. He dubs this ostensible moral apathy the “hallmark of the dismal science.”

Read the rest of this post →

After weeks of being shut down due to safety concerns (via questionable export controls), Anthropic’s Claude Fable is once again accessible to the public with new guardrails that limit its capabilities and revert to inferior models more frequently. OpenAI’s ChatGPT‑5.6 has had a delayed rollout to fulfill a reported request from the White House. These models have been evaluated, withheld from the market, and re-released. But no one knows what made them too dangerous to release, and no one knows what would make them safe enough to release. The obscure and seemingly arbitrary standards of the current evaluation process established by the White House’s executive order on Frontier Artificial Intelligence (AI) show that, while it did not formally introduce an “FDA for AI,” it might have done so in practice. In fact, this regime shares most of the common pitfalls of a vetting regime and, in some cases, might worsen them. 

Read the rest of this post →