MX-5 (NA (MX-5 / Miata / Eunos Roadster))

An American Idea Put to a Japanese Company Bob Hall was a Californian motoring journalist, fluent in self-taught Japanese, whose father had run MGs, Triumphs and Austin-Healeys. Hagerty places the decisive conversation in 1978, when Hall was at AutoWeek and met Kenichi Yamamoto, then head of Mazda research and development, and suggested recasting the rear-drive 323 as a two-seat roadster. Yamamoto was noncommittal. In 1981 he asked Hall to study it properly; Hall joined Mazda's Californian research organisation as a product planner and argued in an early-1982 position paper that recycling existing components was the only way to make the numbers work. The programme ran off-line, and the American position was that it should be a front-engine, rear-drive roadster or nothing at all.

Duo 101, V705 and the Elan on the Bench From 1983 the concept became an internal competition between the Californian studio and the Tokyo team. At the first judging in 1984 California's front-engine rear-drive proposal, Duo 101, beat Tokyo's front-drive and mid-engine alternatives, the mid-engine layout having struggled against noise and vibration targets. Mazda's own account confirms the reasoning: the linear responses of the original lightweight sports cars were judged unobtainable without rear drive, and a new powertrain accepted rather than compromise. A running prototype, V705, was completed by the British consultancy IAD in August 1985 with a glassfibre body over 323 mechanicals. Production approval brought the codename P729 and Toshihiko Hirai as chief engineer. Mark Jordan drew the exterior in California and Masao Yagi pushed the styling further towards the Lotus Elan; Tom Matano and Koichi Hayashi settled the detail once the project moved back to Japan. The Elan was not a vague influence - it was the reference object.

Jinba Ittai and the Power Plant Frame Mazda's design principle was jinba ittai, horse and rider as one. The engineering that carries it is mostly unglamorous: double wishbones at all four corners, near-equal weight distribution, a five-speed gearbox lifted from the rear-drive 929/Luce with a very short throw, and an aluminium power plant frame bolting engine and gearbox rigidly to the differential housing so that driveline movement is taken out of the response. Mazda still calls the frame unique to its sports cars and essential to their directness, while conceding it is awkward to assemble on a mixed line. The car reached the United States in May 1989 as a 1990 model and Japan on 1 September 1989. Autocar recorded 9.1 seconds to 60 mph and 114 mph, called the gearchange the car's real advantage and scored it nine out of ten.

The 1994 Rebuild For the 1994 model year the car was reworked around new side-impact legislation. The 1,839 cc BP-ZE arrived at 96 kW (129 bhp), rising to 99 kW (133 bhp) for 1996; brakes grew, and the shell was braced, most visibly by a track bar between the seatbelt towers but also through both subframes. Dual airbags went into a redesigned dashboard. The car gained weight and, on the published figures, very little speed. In some markets, Europe among them, the 1.6 was kept on below the 1.8 as the cheap option in reduced-output form - by how much is genuinely unsettled, with published figures ranging from 89 bhp to 110 bhp.

Three Names and the End of the Run In Japan the car was sold as the Eunos Roadster, Eunos being a separate premium channel Mazda was building in the manner of Acura or Lexus; in North America it was the Mazda Miata, elsewhere the MX-5. Japanese specials came thick and fast, along with M2 Corporation's low-volume conversions, and Britain received a long series of named editions ending with the Berkeley in 1998. Mainstream production closed in 1997, replaced by the NB, which kept the layout and dropped the pop-up headlamps.

Two faults decide what an NA is worth. The first is rust: sills, rear arches and floors, developing from the inside, expensive to put right properly and frequently disguised. LucaCarMods quotes €500 to €1,000 a side for a competent sill repair and over €1,000 where inner sills and quarter panels are involved. The second is the short-nose crankshaft on early 1.6 engines. Miata.net puts the keyway engagement area at 0.045 square inches against 0.087 on the long-nose part, notes that the timing cog acts as a structural member so the small pulley bolt carries both timing and accessory belt loads, and records the change to the large-bolt big-nose crankshaft at US VIN 209447 during 1991. Worn keyways are not repairable; the fix is a later crankshaft assembly, which on a big-nose conversion also means the oil pump and possibly the front cover. Away from those two, the mechanical package is durable, with Hagerty UK reporting that serviced examples pass 200,000 km without drama. The recurring smaller faults are a silted radiator causing head gasket failure, a rattle on start-up from the hydraulic tappets, rear calipers sticking on cars that sit, and hoods that cloud and split, with Classic & Sports Car putting a full replacement near £1,000.

Structure first. The sills are double-skinned and spot-welded and corrode from the inside, where trapped moisture cannot escape, so a clean-looking sill proves very little. Hagerty's UK guide names the bottoms of the rear wings ahead of the wheels as the worst area, because moisture sits behind the panel, followed by inner and outer rear arches, floorpans, the bases and tops of the front wings and the feet of the windscreen pillars. Classic & Sports Car adds that sills are rarely repaired properly and that clear drain channels are the best available proxy for an attentive owner. Panels welded over existing rust are common and leave the corrosion running underneath. Then the crankshaft. On early 1.6 cars the pulley nose is short and the small pulley bolt carries both timing and accessory loads; the MX-5 Owners Club identifies the short-nose crank by four slots in the pulley against eight on the later part. Look for a wobbling pulley, ferrous powder around the bolt or key, a bent or rusted key, and an engine running poorly because the timing has crept. Beyond that, check the radiator is not silted; check the hood for clouding and rot; check the rear calipers on a car that has stood. Establish market origin from the rear number plate recess, rectangular on European cars and square on a Eunos, and from the VIN prefix - JMZ European, JM0 Australian, JM1 North American - with Eunos cars carrying a shorter Japanese chassis number instead.

As of August 2026, classic.com's NA sub-markets separate more usefully than any single aggregate figure. The base-model page shows a $11,181 benchmark and $11,185 average on a rising trend, with a lowest recorded sale of $2,475 in April 2022 and a $31,998 asking price on a 1996 example - an asking price, not a result. The M-Edition runs higher at a $14,783 benchmark and $14,518 average, also rising, over recorded results from $3,500 in October 2023 to $46,000 for a modified 1997 car in November 2025. The 1993 Limited Edition is the outlier, at a $12,751 benchmark against a $13,352 average on a declining trend, between $7,025 in August 2023 and $13,000 in March 2026. In the United Kingdom, The Classic Valuer reports a median of £9,118 across 850 tracked auction entries at a 91 per cent sell-through rate, with a high of £37,980 in March 2025 and a low of £500 at Mathewsons in June 2024. Its DVLA extract puts 7,298 cars licensed against 13,253 registered but off the road - the clearest indication of how many survivors are projects.