美国零售业：K型经济下的“黄金时代”与“至暗时刻”——全美渠道店及硬线行业研究

Best of Times, Worst of Times: US Retail in a K-shaped economy “It was the best of times, it was the worst of times”. The famous opening line of CharlesDickens’ ‘A Tale of Two Cities’ captures a period of contradiction: the French aristocracyindulging in opulence and privilege, the peasantry experiencing poverty and oppression.While 2026 is thankfully not 18thcentury France, shifting class dynamics remain moretopical than ever. In this note, we examine the US K-shaped economy and implications forBroadlines & Hardlines retailers. Zhihan Ma, CFA+1 917 344 8303zhihan.ma@bernsteinsg.com Jeremy Miles, CFA+1 917 344 8370jeremy.miles@bernsteinsg.com In the US, the COVID period has given rise to an increasingly K-shaped economy. Priorto COVID, the growth in retail spending for different income groups had largely beenclustered. Each group moved together, albeit from a different basis. Since COVID, therehas been a notable decoupling of spend growth by income group. The difference is evenmore stark by education level. Between 2018 and mid-2024, the lowest income group hadgrown spend by ~8% on a cumulative basis, adjusted for inflation, vs. the highest incomegroup growing at nearly double the rate (~17%). It’s been a downward spiral for the low income cohort. Low income consumers are cashconstrained and need to prioritize everyday essentials - food at home, housing, healthcare- with less bandwidth for entertainment, education, and savings. This was already the casebefore COVID in 2019. Since then, the gap has widened further. Given significant inflation inhousing, energy and everyday essentials categories post COVID, the low income cohort hascome under greater inflationary pressure in recent years. This has weighed on their ability tosave and made them more susceptible to inflation, furthering the downward spiral. If persistent inflationary pressure continues to exacerbate a K-shaped economy, whatdoes it mean for retailers? Inflation is generally good news for retailers on the top line.This is especially the case for value oriented retailers. When inflation goes up and otherretailers take price, value-oriented ones remain cheaper. We seeWMT and COSTas thebest positioned to gain share in an environment with persistent inflation.Dollar storescould also benefit from higher income consumers seeking value and trading down, as wehave seen over the last 12 months. Meanwhile, inflationary pressure in essentials categories has weighed on discretionaryspending. Walmart US, Target, and Costco have each seen staples mix up by +679bps,+770bps, and +118bps respectively since 2019 (as % net sales). For more discretionary-oriented retailers (e.g., TGT), this has had a negative margin mix impact. In the near term,we expect inflationary pressure to persist, which will reinforce the downward spiral for lowincome consumers as they continue to prioritize everyday purchases. Over time, as inflationnormalizes and consumer spending power recovers, we see a path for discretionaryspending to recover, led by smaller ticket items and higher income consumers first.Retailers with more discretionary exposure in our coverage (e.g.,FIVE, DLTR, TGT) arebetter positioned in an environment when discretionary mix recovers. BERNSTEIN TICKER TABLE INVESTMENT IMPLICATIONS We rate COST, DG, WMT, and LOW Outperform. We rate DLTR, FIVE, TGT, and HD Market-Perform. DETAILS A K-SHAPED TRAJECTORY - DECOUPLED RETAIL SPEND BY COHORT SINCE COVID-19 PANDEMIC In US Retail, the K-shaped economy is not a new phenomenon. Rather, its manifestation has been visible in data and analysesby major economists for years now. The COVID period coincided with an uncoupling of retail sales growth for differentincome cohorts, where they had previously grown together in the years prior. This is the result of a confluence of factors - theinitial market shock, layoffs and furlough, lockdowns, stimulus and wealth transfers (e.g., SNAP), post-COVID inflation, andgovernment policy. These are all structural, and will take time to reverse. The most obvious way the economy will normalize isthrough disinflation, and the return of purchasing power. It could also normalize “from the top” - through a market collapse andwealth destruction - but that is not desirable. Analysis by Hacıoğlu Hoke et al. (2024) of the Federal Reserve shows, using Numerator data, how retail spending has grown offof the 2018 base (Exhibit 1 & Exhibit 2).1The trend is clear; prior to COVID, the growth in spending for different income groupshad largely been clustered. Each group moved together, albeit from a different basis. Since COVID, there has been a decouplingof spend growth by income group. The difference is even more stark by education level. Between 2018 and mid-2024, thelowest income group has grown spend by ~8% adjusted for inflation. That is slightly above the rate of US population growthon a compounded basis (~1%). In fact, the retail spend of the low income group barely grew in real terms on a cum