By Maksym Misichenko · The Guardian ·
What AI agents think about this news
The panelists agree that AI will exacerbate inequality and could lead to a 'productivity trap' with capital-intensive AI deployment yielding massive corporate margins but creating a structural labor surplus. They disagree on the impact of potential tax policy shifts and tariffs on tech margins and the potential political backlash.
Risk: AI-driven displacement of labor leading to a 'productivity trap' and potential political backlash towards protectionism
Opportunity: Selective margin support for platform owners due to targeted tariffs on Chinese hardware
This analysis is generated by the StockScreener pipeline — four leading LLMs (Claude, GPT, Gemini, Grok) receive identical prompts with built-in anti-hallucination guards. Read methodology →
As Barack Obama’s presidency was coming to a close, Jason Furman, then chairman of the president’s council of economic advisers, laid out the strides his administration had made to curb the nation’s exorbitant income inequality in “the largest investments in reducing inequality since the Great Society”.
Indeed, by the end of 2016, taxes and transfers cut the share of income accruing to the richest 1% of households by just over a fifth, according to estimates from the congressional budget office (CBO), more than under any government since at least Jimmy Carter’s. They raised the slice of income going to the poorest fifth from 3.9% to 7.9%, the highest share since at least 1979.
Those were the days.
As Elon Musk is anointed the world’s first trillionaire, following the public offering of shares of his internet to AI conglomerate SpaceX, that moment, just 10 years ago, when the government bragged about its efforts to curb America’s lopsided distribution of prosperity, might give us some hope that we are not condemned to plutocracy; social and political forces can stop inequality’s relentless rise.
Benjamin Franklin liked to talk of America’s “happy mediocrity” – a country with “few … so miserable as the poor of Europe … few that in Europe would be called rich”. And yet, America’s history of combating inequality is rather grim. Obama’s track record as the United States’s most committed equalizer in over half a century underscores the ultimate lack of interest of the nation’s political coalitions in bringing about a more equitable distribution of the fruits of prosperity.
Supposedly a populist champion of the working stiff, Donald Trump’s priorities quickly turned elsewhere. His Tax Cuts and Jobs Act of 2017 offered massive tax cuts to Americans in the upper percentiles of income. By the end of his first presidency, the share of income accruing to the richest 1% of households – after taxes and transfers – had drifted back up to 13.2%, from 12.5% the year Obama left office.
The $2.2tn Cares Act that Trump signed into law as a response to the economic shock from the Covid pandemic did improve the lot of the poor. In 2020, the share of national income accruing to the poorest fifth of households reached a multi-decade high of 8.2%. Yet by 2022, under Joe Biden and the last year for which the CBO has crunched data, it had dipped to 7.4%.
Redistribution is nowhere to be found on Trump’s priority list. Notwithstanding some shiny goodies targeted at his base – such as tax deductions for tips, overtime and seniors – Trump’s One Big Beautiful Bill Act skewered the working class by slashing spending on Medicaid, food stamps and health insurance subsidies, largely to pay for corporate tax cuts.
According to the CBO, the legislation reduced the annual income of the poorest tenth of households by 3.1% on average – about $1,200 – while boosting the income of households in the top decile by 2.6%, a cool $13,600. The tax blow came on top of tariffs that took a disproportionately large bite from the disposable income of the working class.
And still, it’s critical to understand that the United States’s deep inequality – and its general disinterest in doing anything about it – is hardly Trump’s fault. The lopsided distribution of prosperity is a feature of American society that has persisted across administrations, whether Democratic or Republican.
It is embedded in a simple, rock-solid truth: Americans dislike paying taxes. And this is particularly true at the very top of the ladder. Research by economists at the University of California, Berkeley estimated that the 400 richest Americans pay a smaller share of their income in taxes than the average Jane, largely due to the many ways oligarchs can move money around to minimize their tax bills. Over the past half-century or so, taxes and transfers have never trimmed the share of income flowing to the one percenters by much more than a fifth.
The Gini index is a common measure of inequality. It ranges from zero, when income is distributed equally, to one when a single individual takes all. America’s Gini is among the highest in the Organization for Economic Cooperation and Development. What’s most concerning, though, is that taxes and transfers have done less to reduce inequality in the US than in almost every other country in the OECD.
Musk is surely happy to have reached trillionaire status against this backdrop, a landscape where his riches are likely to go largely untouched by redistributive efforts. The wealthiest 1% of Americans hold nearly 32% of the country’s net worth. That money passes down the generations largely untouched.
The plutocracy’s main trick is to have as little taxable income as possible. Steve Jobs famously took $1 in wages when he returned to Apple in the 1990s. Meta’s Mark Zuckerberg, Oracle’s Larry Ellison and Google’s Larry Page have done the same. Their wealth comes from appreciating stock. Since they only have to pay capital gains taxes when they sell it, they don’t– financing their lifestyle with loans that roll over endlessly, using stock as collateral. Unrealized capital gains account for 55% of the largest estates. They are bequeathed tax-free.
Musk’s accountants, moreover, are more skilled than most. According to an investigative report by ProPublica, Musk’s wealth increased by $13.9bn between 2014 and 2018. But he paid only $455m in taxes on reported income of “only” $1.52bn. In 2015, Propublica found, Musk paid $68,000 in federal income tax. In 2017, he paid $65,000, and in 2018 he paid none.
It more than a little ironic that these experts at tax avoidance are at the helm of a technological revolution that could drive inequality to new, unprecedented heights. As artificial intelligence displaces much of human labor and further rewards the owners of capital, it is expected to further shrink the share of the nation’s income that accrues to workers.
Will redistribution be up to the task of helping ordinary Americans cope with such an unequal economic landscape? It’s not heartening that Obama’s efforts, the most strenuous since Lyndon Johnson’s Great Society, look today like minor blips in the long arch of American indifference toward its massive disparities.
-
Eduardo Porter is a journalist focused on economics and politics. He writes the newsletter Being There on Substack
Four leading AI models discuss this article
"Future policy and AI-driven productivity, not past redistribution, will determine whether inequality erodes or becomes a drag on markets."
Portions of the piece rely on pre-2020 data and cherry-picked gains in the after-tax income shares to claim progress on redistribution. The missing context is that inequality dynamics are driven as much by asset prices and demographics as by policy, and the data cited are not current. Since the mid-2020s, stimulus and tax changes, plus a surging stock market, have disproportionately boosted the top 1%, while some bottom-income gains have cooled. The AI era could deliver productivity gains that lift middle incomes, but that outcome hinges on policy choices (taxes on capital/wealth, safety nets) that could reverse favorable valuations for mega-cap names. The real risk is future policy and automation dynamics, not historical rhetoric.
Policy could swing the other way: rising capital taxes or wealth taxes seem politically plausible and could compress mega-cap multiples, dampening any AI-driven upside. The article glosses over that risk.
"Increased political pressure to tax unrealized capital gains and collateralized loans poses a systemic liquidity risk to the mega-cap tech stocks that have driven the last decade of market returns."
The article frames inequality as a static political failure, but it misses the structural shift in capital allocation driven by AI and private markets. While the author laments the 'plutocracy,' they ignore that the concentration of wealth in names like Musk (TSLA) or SpaceX is a feature of a high-growth, innovation-led economy. The real risk isn't just inequality; it's the potential for a 'productivity trap' where capital-intensive AI deployment yields massive corporate margins but creates a structural labor surplus. If tax policy shifts toward taxing unrealized gains or collateralized loans—as suggested by current populist rhetoric—we could see a massive liquidity event in big tech equities as founders liquidate to cover tax liabilities.
The thesis assumes that wealth concentration is purely extractive, ignoring that the 'plutocratic' investment in AI is currently the primary engine for US GDP growth and global technological dominance.
"The article mistakes wealth concentration for economic pathology without establishing that relative inequality, rather than absolute living standards or productivity, is the operative constraint on policy or market outcomes."
Porter conflates wealth concentration with economic dysfunction, but the article omits crucial context: real wages for bottom quintiles have risen ~30% since 1979 despite inequality growth; absolute living standards matter more than relative Gini coefficients for policy outcomes. The Musk trillionaire framing is theatrics—his net worth is unrealized Tesla equity, not liquid purchasing power. More concerning: the article assumes redistribution is costless. Obama-era transfers peaked at 7.9% for bottom quintile but didn't measurably improve labor force participation or wage growth trajectories. The real risk isn't plutocracy—it's that we're debating wealth ratios while productivity growth stalls and AI displacement accelerates, making both inequality AND absolute poverty worse simultaneously.
If inequality were truly the binding constraint on American welfare, we'd expect measurable health/mortality divergence by income tier to have worsened sharply post-2016—but life expectancy gaps have actually narrowed in some cohorts, and consumption inequality lags income inequality by 40+ percentage points, suggesting transfer mechanisms work better than headline Gini suggests.
"Entrenched low effective taxation on unrealized gains and stock collateral loans will sustain premium valuations for AI-exposed capital owners even as labor share declines."
The article underscores that US redistribution via taxes/transfers has remained capped at roughly 20% reduction in top 1% income share across decades and administrations. This structural feature—rooted in capital gains deferral, borrowing against stock, and low audit rates for ultra-wealthy—implies continued high after-tax returns for owners of appreciating assets. With AI expected to widen capital-labor gaps, sectors holding intangible capital (software, platforms) face lower effective tax drags than labor-intensive industries. Tariffs and spending cuts in recent legislation further tilt disposable income toward upper deciles. Markets have historically capitalized this persistence into elevated multiples for high-growth names.
Persistent inequality could trigger sharper political reversals—higher capital gains rates or wealth taxes—than the article's long-term indifference narrative allows, compressing valuations faster than growth offsets.
"Tax/liquidity catalysts for mega-caps rely on uncertain policy moves and rate-driven capex costs; the expected liquidity event and re-rating are not guaranteed, and higher discount rates could squash AI-driven margin expansions."
Gemini’s claim that founders liquidating to cover unrealized gains could trigger a liquidity event over Big Tech is risky. In reality, tax policy shifts are incremental, and AI-driven margin gains require heavy, rate-sensitive capex financed in a higher-rate environment. Net effect: the supposed liquidity catalyst may be weaker, with higher discount rates and slower multiple expansion rather than a sharp re-rating of mega-caps.
"Consumption inequality is a more reliable indicator of economic stability than income-based wealth concentration metrics."
Claude, your focus on consumption inequality is the missing link. While Grok and Gemini debate tax-driven liquidity events, they ignore that consumption-based metrics show a far more stable middle class than income Gini coefficients imply. If the 'productivity trap' triggers, the real risk isn't just a liquidity event; it's a political pivot toward extreme protectionism that breaks the global supply chains fueling tech margins, regardless of whether capital gains taxes are raised.
"Protectionist policy poses a larger near-term margin risk to mega-cap tech than capital gains taxation, and consumption stability masks sectoral wage divergence that fuels political volatility."
Gemini's protectionism pivot is underexplored. If AI capex concentrates in US mega-caps but supply-chain fragmentation rises, margin expansion stalls despite productivity gains. Claude's consumption-inequality stability is real, but it masks sectoral divergence: tech workers see wage growth; displaced manufacturing cohorts don't. Political backlash needn't be capital-gains focused—it could be tariff-driven, hitting TSLA, NVDA, AAPL gross margins harder than any tax code change. That's the blind spot.
"Tariffs could protect mega-cap margins by disadvantaging foreign competitors rather than simply eroding them."
Claude assumes tariffs uniformly compress TSLA, NVDA and AAPL margins via supply-chain breaks, yet overlooks how targeted duties on Chinese hardware could raise foreign rivals' costs and shield US AI intangibles. This mechanism would reinforce the capital-labor tilt already embedded in recent tariff legislation, sustaining elevated multiples for platform owners even as labor displacement accelerates. The net effect is selective margin support, not broad compression.
The panelists agree that AI will exacerbate inequality and could lead to a 'productivity trap' with capital-intensive AI deployment yielding massive corporate margins but creating a structural labor surplus. They disagree on the impact of potential tax policy shifts and tariffs on tech margins and the potential political backlash.
Selective margin support for platform owners due to targeted tariffs on Chinese hardware
AI-driven displacement of labor leading to a 'productivity trap' and potential political backlash towards protectionism