Shunto 2025 Through a Critical Lens: Why Japan's Best Wage Round Since 1991 Was Called a Settlement 'Lacking Momentum'
The 2025 round of Japan's coordinated spring wage negotiations, known as the shunto, closed with an average increase of 5.25% across the 5,162 companies captured in the final tally of the Japanese Trade Union Confederation, Rengo — the strongest settlement since 1991 and the second consecutive year of outsized gains. Yet the same round that produced a 34-year high also attracted one of the sharpest critical readings in recent memory. In an April column, Takahide Kiuchi of the Nomura Research Institute argued that the 2025 wage hikes "lack momentum," tracking inflation rather than restoring the purchasing power households lost in earlier years. This analysis sets the final tally against that critique and asks what the distance between the two readings means for the Bank of Japan, for household consumption, and for the country's long exit from the lost decades.
The final tally: 5.25% and a 34-year high
The definitive number for the 2025 shunto came with Rengo's final tally, reported by The Japan Times on July 4, 2025. Across the 5,162 companies whose settlements the confederation aggregates — a base representing roughly seven million union members — the average negotiated wage increase settled at 5.25%. That is the highest final figure since 1991, and it confirms that the 2024 round, which had already broken a multi-decade ceiling, was not a one-off but the second year of a genuinely re-accelerating wage cycle.
The final number is nonetheless lower than the preliminary count Rengo published in March, which stood at 5.46%. That gap is normal: early tallies are weighted toward the large, well-organised exporters that settle first and settle high, while later-reporting small and mid-sized firms drag the average down as their results are folded in. The direction of travel between the two counts therefore carries information of its own about where the round's strength was concentrated.
The key figures of the 2025 round, as consolidated in the final tally, are as follows:
- average negotiated increase across 5,162 companies — 5.25%, the highest since 1991;
- preliminary March tally — 5.46%, later revised down as smaller firms reported;
- increase for workers at smaller firms — 4.65%, up from 4.45% a year earlier;
- coverage of the tally — roughly seven million Rengo members.
The structure of the round explains why the numbers arrive in waves. Shunto is not a single negotiation but a sequence: unions file coordinated demands at the start of the year, the largest companies answer first in March, and successive response rounds through the spring bring in smaller employers, with Rengo aggregating the results at each stage and publishing a final tally once the long tail has reported. The March figure of 5.46% is therefore a snapshot of the front of the round, the third-round count of 5.42% a snapshot of its body, and the July figure of 5.25% the settled whole. Each of the three numbers answers a slightly different question, and much of the apparent disagreement between celebratory and critical readings dissolves once the question behind each number is made explicit.
Read on its own, this is a triumph for the union movement. A 5.25% average settlement in an economy that spent three decades expecting roughly one percent is a structural event, not a rounding error. It is also the number the Bank of Japan and the government cite as evidence that the wage-price cycle has finally turned. The critical reading, however, begins precisely where the headline number ends.
The April counter-reading: a record that "lacks momentum"
Three months before the final tally, Takahide Kiuchi, an economist at the Nomura Research Institute and a former Bank of Japan executive, published a column on April 11, 2025 under a blunt heading: the shunto wage hikes, he wrote, lack momentum. His argument was not that the increases were small in nominal terms — by historical standards they were large — but that they were small relative to what Japanese households had already lost and relative to the inflation they were meant to outrun.
Kiuchi's evidence rested on the third-round tally of the round, the count that captures the bulk of settled unions. Including the regular, seniority-driven salary bumps that Japanese workers receive automatically each year, the third-round average increase stood at 5.42%. Stripping those automatic increments out and looking only at base pay — the component that actually represents a negotiated raise — the figure fell to 3.82%. Both numbers exceeded the comparable 2024 readings of 5.24% and 3.63%, so the round was an improvement year on year. But the improvement, in Kiuchi's framing, was one of degree rather than of kind.
The base-pay lens
The distinction between the all-in figure and the base-pay figure is the analytical hinge of the entire debate. Japan's traditional wage system grants workers an automatic annual increment as they age and accumulate seniority; that increment arrives whether or not management concedes anything at the negotiating table. The base-pay component, known as the base-up, is the only part of a settlement that reflects a genuine transfer from profits to labour. A round can therefore post a large headline number while the negotiated core of that number remains modest.
On that lens the 2025 round looks less like a breakout and more like a continuation. A base-pay increase of 3.82% against 3.63% a year earlier is real progress, but it is progress measured in tenths of a percentage point, and it sits below the pace of the inflation that eroded real incomes in the preceding years. This is the sense in which the hikes, in Kiuchi's words, were "simply following the high rate of inflation" rather than leading it.
The SME gap that keeps widening
The second pillar of the critical reading is the divide between large and small employers. In the third-round tally, unions at large firms — those with 300 or more Rengo members — settled at an average of 5.44%, while unions at small and mid-sized firms, with fewer than 300 members, settled at 5.00%. The final tally tells the same story from a different angle: workers at smaller firms received 4.65%, up from 4.45% a year earlier but still well short of the 5.25% average.
The gap matters because smaller firms employ the majority of Japan's workforce. A wage cycle that is strong at the large-exporter core and weaker at the small-firm periphery lifts the national average without lifting the median household in proportion. Kiuchi's conclusion was that the wage-level gap between large and small firms was "only expanding further," a judgement that reframes the record headline as a story of uneven diffusion rather than broad-based recovery.
The reasons for the gap are structural, which is why it persists. At a small firm, a wage increase consumes a share of revenue that a large exporter can pass into the price of an export contract; the small firm has less margin cushion and less bargaining power over its own customers. As long as the prices it charges lag its wage bill, each new settlement is harder than the last — and the diffusion of the record into the depth of the economy proceeds more slowly than the record itself.
Wages chasing prices: the real-wage arithmetic
The decisive test of any wage round is not the nominal percentage but what it does to purchasing power after prices. On that test the spring of 2025 was still a period of decline. Real wages — nominal pay adjusted for consumer prices — fell 2.8% in January and 1.2% in February, according to the labour ministry data cited in the April commentary. A worker whose pay rose in April had, until that point, been running behind prices for months.
The arithmetic of the recovery follows a specific sequence, and the sequence explains why a record nominal round can coexist with a continued real decline:
- inflation ran ahead of wage settlements through 2024 and into early 2025, pushing real wages negative;
- the 2025 settlements, agreed in spring, only begin to reach pay packets from April onward;
- because the increases merely tracked the prevailing inflation rate, real wages stayed negative until the new pay overtook the price index;
- on Kiuchi's estimate, that crossover arrives only around the middle of the year and into the summer, leaving the first months of 2025 in real decline.
Looking at the full year, the column expected the 2025 average wage increase including regular rises to come in around 5.3%, against 5.1% in 2024, while total cash earnings — the broader measure of nominal pay — were projected to grow by roughly 3.1%. Neither figure is a real-terms gain of the size the headline 5.25% implies. The column's verdict was therefore that the round, however large in nominal terms, was "inadequate" to recoup the real income households had already surrendered.
Equally important is what these estimates do not promise: none of them offers a rapid recoupment of the losses of the inflation years. The column explicitly calls the hikes inadequate to return to households the real income already lost — in its optics the 2025 round works as partial compensation for current price pressure, not as restitution of the past. For a household the difference is fundamental: the first improves statistics and sentiment, the second changes the budget.
What the gap means for the Bank of Japan
For the Bank of Japan the shunto is not a labour-market curiosity but a policy input. The central bank's case for normalising policy away from decades of ultra-easy money rests on the existence of a virtuous wage-price cycle: firms raise wages, households spend, prices rise on demand rather than on imported cost, and higher prices in turn justify higher wages. A round in which wages merely follow prices — rather than lead them — supplies weaker evidence for that cycle than the headline number suggests.
This is why the composition of the settlement matters as much as its size. Base-pay increases are the component most likely to persist into future rounds and most likely to feed durable domestic demand; seniority increments expire with the cohort that receives them. Likewise, diffusion to small firms is the test of whether the cycle is economy-wide or confined to the export core. On both counts the 2025 round was an improvement on 2024 but not a transformation, which is precisely the nuance the critical reading insists upon.
The practical implication is one of patience rather than reversal. A wage round that tracks inflation still removes some of the deflationary drag that defined the lost decades, and a second consecutive strong round keeps the cycle alive. But it does not, on its own, demonstrate the self-sustaining wage-price dynamic the central bank needs to see before it can treat normalisation as settled.
Two developments would materially strengthen the evidence base. The first is a base-pay component that holds near or above the 2025 level in the following round, which would show that the negotiated core of the cycle is durable rather than a one-off catch-up payment. The second is a narrowing, rather than a widening, of the gap between the large-firm and small-firm tiers, which would show that the cycle is travelling through the supply chains and local economies where most workers are paid. Absent those, the wage data will continue to support normalisation only gradually, and the critical reading will retain its force.
Consumption and the long exit from the lost decades
Household consumption is where the wage round either becomes an economic recovery or remains a statistical one. For a household, the settlement percentage matters less than the distance between the pay packet and the checkout: while real wages fall, every nominal gain is absorbed by prices before it can become spending. The April commentary's judgement that the increases were inadequate to recoup past real-income losses therefore carries a direct consumption implication — households must first recover the ground lost during the inflation years before the wage cycle can express itself in demand. A real-wage line that stayed negative into the spring, and that only crosses into positive territory around mid-year, implies a consumption recovery that is gradual and back-loaded rather than immediate.
The uneven geography of the round compounds the caution. Because smaller firms — where most workers are employed — settled below the average, the median household's pay rise lagged the celebrated 5.25%. Consumption is driven by the median household, not by the export-sector average. Until the small-firm tier closes the gap, the national settlement figure will overstate the spending power it creates.
Set against the longer horizon, the 2025 round is best read as a milestone inside a transition rather than its conclusion. Two consecutive strong rounds begin to repair three decades of wage stagnation, and a base-pay component approaching four percent is something Japan has not seen in a generation. But the lost decades were defined by a shortfall of real income, and a nominal record that merely tracks prices repairs that shortfall slowly. The exit, on this reading, is under way — and is measured in years, not in a single spring.
The verdict of the two readings
The final tally and the April critique are not in factual conflict; they are two measurements of the same round taken with different instruments. The tally measures what was agreed, and on that measure 2025 was the strongest shunto since 1991. The critique measures what the agreement buys, and on that measure it restores purchasing power only partially and only late in the year.
Holding both readings together yields a coherent picture: a wage cycle that has genuinely re-accelerated at the top, that is diffusing unevenly toward the small-firm majority, and that is outrunning prices only just. The indicators that will decide which reading history endorses are already visible in the round itself — the base-pay component, the large-versus-small gap, and the path of real wages through the second half of 2025. On those three lines, the record 5.25% is a beginning rather than a verdict.
Leave a comment