When the Second Sino-Japanese War flared in 1937, Chinese units flew the Soviet Tupolev SB, a sleek twin-engine bomber whose speed let it strike and escape before Japanese fighters could reach it. For an army used to owning the sky over China, that was intolerable. The response was a formal December 1937 instruction to Kawasaki for a fast light bomber of its own — a machine defined less by what it should carry than by what it should be able to outrun.

The result, the Ki-48, was a clean cantilever mid-wing monoplane with retractable gear and two radial engines. It hit its speed target and, for a year or two, delivered exactly the immunity the Army had wanted. The deeper lesson — that a bomber optimised purely for speed in 1938 would be a slow, fragile target a few years later — would be learned the hard way over Burma and New Guinea.

Japanese and Western light bombers alike leaned on defensive machine guns to survive interception, but the Ki-48 went to war with the bare minimum: three flexible 7.7 mm Type 89 guns, the same rifle-caliber weapons that had armed 1930s aircraft. Against a diving P-38 Lightning or a Spitfire, three hand-held popguns offered almost nothing, and the unarmoured, unprotected fuel system behind them meant a single burst could be fatal.

The Army knew this. The up-armed Ki-48-IIc substituted a 12.7 mm gun in the dorsal mount, and surviving units increasingly flew at night to avoid interception altogether. But the fundamental problem — a fast-bomber philosophy that had traded away defence for speed — could not be patched out of the airframe.

Three Ki-48-II prototypes flew in February 1942 and the variant entered production that April. The Ha-115 engines added power and altitude, the structure was strengthened, and — learning from the Pacific — the II finally gained some armour and partial fuel-tank protection and a larger bombload of up to about 800 kg. One sub-type, the Ki-48-IIb, added under-wing dive brakes to turn the level bomber into a dive-bomber, a mid-life role change forced by battlefield reality.

It was a genuinely better aircraft than the Ki-48-I — but “better” was measured against a 1938 design, and the improvements never closed the gap with the fighters now hunting it. Production of the II ran until October 1944.

There is no dependable public figure for what a Ki-48 cost to buy or to operate. It was a wartime product of Imperial Japan’s state-directed aircraft industry, never sold on an open market, and Japanese wartime procurement costs are poorly documented in Western-accessible sources. No reviewed reference — museum, encyclopedia or specialist history — lists a unit price or a cost-per-flight-hour for the type.

So both the unit-cost and cost-per-flight-hour rows above read “no reliable public figure exists,” and any precise number you see attached to a Ki-48 should be treated with suspicion. This field is kept for consistency with the rest of the encyclopedia, where operating-cost data does exist for later aircraft.