A Census Taken by the Enemy

Before the end of this year the Director of National Intelligence must publish, on a public government website, an estimate of the personal wealth of Xi Jinping and the six other members of the Politburo Standing Committee. The requirement sits inside the FY2026 Intelligence Authorization Act, which became law on December 18, 2025 as part of the annual defense bill. The report has to cover financial assets, business interests, real estate at home and abroad including Hong Kong and Macau, and the proxies and relatives who hold assets on the leaders' behalf. Congress gave the DNI 180 days plus 60-day extensions. ODNI confirmed in June that the report is in progress, which means the first deadline has already come and gone.

This is the second attempt. A 2023 law ordered the same report, and ODNI delivered it in March 2025, eighteen months late and forty pages long. The document conceded that it did not provide a detailed list of any individual's holdings. It leaned on the 2012 New York Times investigation of Wen Jiabao's family, which traced about $2.7 billion in assets, and on 2012 reporting that Xi's extended family had grown rich, and it recorded that Xi's own anti-corruption campaign has investigated and found guilty nearly five million officials. Congress read the report as an evasion and wrote the second mandate with teeth: a baseline, named proxies, nonpublic intelligence where sources allow.

Beijing's reaction to the new law was louder than its reaction to any US tariff. The foreign ministry called it interference in China's internal affairs, which is a strange thing to say about a document that describes bank accounts in Vancouver and Luxembourg. The sensitivity has a simple source. The party's claim to rule rests on the assertion that its leaders serve the people rather than themselves, and that assertion has never been tested against a retirement account statement because the party will not allow one. In 2014 the legal scholar Xu Zhiyong received four years in prison for organizing a campaign whose central demand was that officials disclose their assets. A foreign government is now preparing to publish the figures that Chinese citizens go to prison for requesting.

The Roman Republic treated declared wealth as the foundation of citizenship. Every five years two censors summoned every citizen to declare, under oath, his property down to the last slave and the last iugerum (an area an ox could plow in a day or about 62% of an acre) of land. The declared figure decided everything that mattered: which of the five property classes he voted in, what military equipment he owed the state, what tribute he paid, and whether he qualified for the Senate or the equestrian order. A man who skipped the census was incensus, and the law allowed the state to flog him and sell him into slavery. Rome did not treat concealment of wealth from the state as a private matter. It treated concealment as a crime against the constitution.

The censors also policed the top. When Cato the Elder held the office in 184 BC he assessed jewelry, women's clothing, carriages and expensive slaves at ten times their market value and taxed them at that figure. He struck seven men from the Senate, among them a former consul, and he attached a written mark, the nota censoria, to each name so that posterity knew why. The office carried no army and no court. It carried a list, that was publicly posted. That list was enough to end careers.

That is the frame for what Washington is doing. The CCP has no censor and tolerates none. Rank in the party confers the privilege of being unregistered, and the higher the rank the more complete the privilege. The United States is about to perform the office from outside, for a political class that has spent seventy years making sure nobody could perform it from inside.

Before Washington congratulates itself on becoming Beijing's censor, it should look at its own ledger. Members of Congress file annual financial disclosures and report every trade over $1,000 within 45 days, and a popular line of commentary treats that fact as a trump card: American politicians disclose, Chinese politicians do not, end of argument. The disclosure exists. The question is what it discloses, and what the people who file it have done while filing it.

Start with the best investor in Washington. Nancy Pelosi's household portfolio, traded by her husband Paul and disclosed under her name, returned about 65% in 2023 and about 71% in 2024 by Unusual Whales' estimate, against 24% and 25% for the S&P 500. Capitol Trades puts the cumulative return over the past decade at roughly 816%, with Berkshire Hathaway at about 282% over the same stretch. UW estimates her net worth at more than $640 million as of mid-2026, up from $121 million in 2013. Warren Buffett spent sixty years building the most studied investment record in the world. A congresswoman from San Francisco nearly tripled it in ten, including a well-timed purchase of Nvidia call options in the summer Congress was debating the CHIPS Act. Not shady at all.

The habit is bipartisan. On April 8 and 9, 2025, the day before and the day of the president's announcement pausing his tariffs, Marjorie Taylor Greene bought between $21,000 and $315,000 of stock, Palantir and AMD among the names, and rode the rally that followed. Jared Moskowitz, a Florida Democrat, bought into twenty companies on April 7. Both said an outside adviser placed the trades without consulting them. Greene's portfolio rose about 476% between her arrival in Congress in 2021 and her departure this January. In 2024 the portfolios of Democratic members averaged about 31% and Republican members about 26%, both ahead of a market that returned 25%.

Ilhan Omar is the case where the ledger fails at the reading stage. Her 2024 disclosure, filed in May 2025, listed her husband's stake in a venture capital firm at $5 million to $25 million and his stake in a winery at $1 million to $5 million, for a household total somewhere between $6 million and $30 million. Two years earlier the same couple had reported a maximum of about $158,000, and her 2019 filing showed a negative net worth. She said she was barely worth thousands, her office blamed an accounting discrepancy, and this June's filing showed a potentially negative net worth again. The winery has since closed, and the president called in January for an investigation. I am not interested here in which number is right. I am interested in the fact that the form cannot say.

Then there is the matter of who speaks for whom. Elizabeth Warren has spent fourteen years in the Senate as the tribune of the squeezed middle class and reports a net worth that Forbes put at about $12 million. Bernie Sanders ran two presidential campaigns against the millionaire class and became a millionaire himself after the first one, on book royalties from Our Revolution; estimates now put him near $3 million. Asked about it in 2019, he said that anyone who writes a best-selling book can be a millionaire too, and his stump speech, which for years named millionaires and billionaires, has since narrowed its target to billionaires. Neither figure suggests concealment, and neither is large by the standards of the chamber. Both senators would nonetheless have been registered in the first census class, and both have built careers on speaking for the fifth. Rome's answer to that tension was to publish the class, as a number, next to the name.

The instrument makes all of this hard to read and impossible to audit. Congressional disclosures report assets in bands that widen as the money grows: $1,001 to $15,000 at the bottom, $5 million to $25 million near the top, and over $50 million as the final box. A single filing can support both "worth thirty million" and "not a millionaire" without anyone lying, because the form permits a five-fold range on every line. Nobody at the House Clerk's office checks the figures against a brokerage statement. The filer is the only source. A Roman censor would not have accepted this. The census demanded a figure, in asses, under oath, and the censors assessed the property themselves when a declaration looked light. The result in Washington reads like a census taken by a clerk who was forbidden to leave his desk.

None of this is a partisan complaint. The same bands cover all 535 members, the best-timed trades of the 2025 tariff week came from a Georgia Republican and a Florida Democrat, and the outperformance runs through both caucuses. If the United States believes that a public, verified, proxy-inclusive accounting of a leader's wealth is a legitimate instrument of accountability when the leader sits in Zhongnanhai, it is hard to explain why the standard should relax when the leader sits on Capitol Hill, trades options on the committee's subject matter, and reports the result in a box.

In 70 BC Cicero took the case of the Sicilians against Gaius Verres, who had spent three years as governor stripping the province of its art, its grain and its tax revenue. Verres had powerful friends, a famous advocate, and a plan to delay the trial into the following year when a friendlier court would sit. Cicero gathered his evidence in fifty days, skipped the customary long opening speech, and put the witnesses straight on the stand. After the first session Verres fled to Marseille. The trial never reached a verdict.

Cicero then published the five speeches he had prepared and never delivered. They ran to several hundred pages of itemized theft: the statues taken from which temple, the bribes paid by which town, the ships sold to pirates. The Verrines became a school text within a generation, and the name Verres became the Latin shorthand for a plunderer. There was no conviction and no sentence. There was a public record, and it did the work the court could not do.

Wen Jiabao is the modern Verres. In 2012 the New York Times mapped the holdings of his wife, his son, his mother and his brother-in-law through layered shell companies to a figure of about $2.7 billion. Wen was never charged. He was never investigated. He retired on schedule in 2013 with every honor the party confers. His standing as the clean reformer, Grandpa Wen, did not survive the week the story ran, and it has not recovered in the thirteen years since. The party could block the newspaper's website. It could not make the document unexist.

The lesson Beijing drew from Wen is the lesson Cicero taught: documentation outlives adjudication. The DNI report won't convict anybody. It won't need to. It needs only to exist, in a form that can be cited, copied and translated, and from then on every speech about serving the people competes with a table of assets that the speaker cannot explain. The Sicilians could not have prosecuted Verres themselves, because the courts belonged to his class. The Chinese people cannot audit the Standing Committee themselves, because the auditors belong to the party. In both cases the accounting must come from outside the structure that produced the men being counted.

Rome offers a warning alongside the precedent. In 82 BC Sulla posted lists of names in the Forum. The men named forfeited their property and their lives, and informers who brought in a head collected a bounty. The first list held 80 names, the later ones thousands, and contemporaries noticed that the lists grew as the treasury's needs grew. Men were proscribed for their villas. Caesar did something subtler a generation later: as consul in 59 BC he ordered the daily proceedings of the Senate published, so that every senator who voted against him did so on the record and in front of the mob. Transparency in Roman hands was a weapon at least as often as it was a reform, and which one it was depended on who held the pen.

Beijing will call the DNI report a proscription list, and the charge is not empty. The author is an adversary with an interest in the outcome. The first report was thin, late and visibly reluctant. The second is already past its deadline, and ODNI has every bureaucratic reason to produce another survey of press clippings rather than the names and numbers Congress asked for. A list compiled by a hostile intelligence service, with sources it cannot reveal and motives it does not need to explain, is not the Roman census. It is closer to Caesar's acta, published to embarrass a rival rather than to govern a republic.

That does not make the report worthless. It makes the report's value contingent on its specificity. A document that names the proxy, the trust, the address in Vancouver and the sum is Cicero. A document that says the leadership has amassed significant wealth through family members is a press release. Congress wrote the second law because it got the press release the first time. Whether it gets Cicero this time is the thing to watch in December. Either way, it's a hell of a bargaining chip.

Tacitus wrote that the most corrupt state has the most laws. Xi Jinping's anti-corruption campaign has investigated and found guilty nearly five million officials since 2012, by ODNI's count, and the party presents that number as proof of its virtue. It is also proof of the scale of the problem, and it has a shape: five million convictions below the Politburo and zero disclosures above it. The campaign audits downward. Nothing in the system audits up. That's a thick line.

The Roman census audited up. That was its point, and it is why Cato could value a consul's silver and Cicero could publish a governor's thefts. The United States is about to do for China's Standing Committee what the Republic did for its own senators, and it should notice, while doing so, that it files its own senators' fortunes in bands and never checks the numbers. The report is due by December 31. If it arrives with names and numbers, it will be the most consequential forty pages Washington has published about China in a decade. If it arrives as another summary, Congress will have learned what the Sicilians knew, which is that passing the law is the easy half and the republic still has to find a Cicero willing to read the ledger aloud.