Japan's slowdown is a capex problem, and it does not clear until the oil does

Japan's GDP growth slowed to roughly 0.3% annualized in the April-June quarter, down from a revised 1.8% in the first three months of 2026, according to cryptobriefing.com ^[1][2][3]. What matters for anyone allocating capital into the world's fourth-largest economy is the composition: capital expenditure was flat at 0% year-on-year in Q1 and business investment contracted 0.7% in the same period, while consumer spending was robust enough to carry the earlier headline ^{[4][5][6][16]}.

The deceleration is about 1.5 percentage points, leaving the second quarter running at roughly a sixth of the first quarter's pace ^[1][2]. The report says the weak-investment trend carried into Q2 with little sign of improvement ^[7]. Note also that the source gives two investment numbers for the same window without reconciling them, a flat 0% for capex and a 0.7% contraction in business investment, so treat the exact magnitude as soft and the sign as the signal ^[3].

That distinction changes what you would expect to fix it. A consumer-led slowdown responds to wage growth, transfers and confidence. An investment-led one responds to a firm's view of input costs and end demand over the life of the asset, and both of those inputs currently run through the Middle East. Japan imports approximately 90% of its crude oil from the region ^[8]. The ongoing conflict has kept crude prices elevated, squeezing manufacturer margins and raising household costs ^[9], and exports to the Middle East fell 17.6% in the first half of 2026 against the same period last year ^[10]. Higher input prices and a shrinking regional order book are the two things most likely to make a plant manager defer a line rather than approve it.

Tokyo's response so far is subsidies to cushion energy costs for households and small businesses ^[11]. That addresses the cost pass-through, which is the visible political problem. It does not restore the forward visibility a capex committee needs, and it is not designed to.

The policy bind is the more consequential part. The Bank of Japan has been gradually normalizing after decades of ultra-loose settings, and cryptobriefing.com notes that a weaker growth backdrop makes further tightening harder to justify ^[12][13]. So the yen and the rate path are now partly a function of oil prices that no one in Tokyo sets.

There is one live counterweight. Business sentiment surveys show pockets of resilience in sectors tied to the global AI buildout, with Japanese semiconductor equipment makers and advanced materials companies still seeing strong demand, which the report suggests could produce targeted capex increases even while broad investment stays subdued ^[14][15]. That is a narrow, exportable slice of the economy rather than a substitute for domestic reinvestment.

The report frames the drop as steep enough to warrant concern but not a recession signal ^[17]. Watch whether Q2 capex confirms the Q1 stall rather than reversing it, whether Middle East exports stabilise in the second half after the 17.6% first-half fall ^[10], whether the energy subsidies are extended or allowed to lapse ^[11], and whether AI-linked demand shows up as approved Japanese capex rather than order-book commentary ^[14][15].