The 90s Sports Car Explosion
Running a successful car company is all about risk: calculating it and minimising it. This is why when one does something different, quite soon they’re all doing something different. Ie. The same thing.
In modern times this means SUVs and cross overs. Lots of them. One company pioneers, the rest see it working and decide as quickly as possible to get in on the act.
In the 1990s it wasn’t about sitting high in a big box, it was about its polar opposite: low, small sports cars.
Low, small sports cars weren’t new. British Leyland specialised in them until they killed the market by steadfastly trying to sell the market what it had been selling to the market for 20 years. And then by offering customers the TR7.
In the 80s those sports car customers migrated to hot hatches and car companies decided nobody wanted sports cars. Except Mazda. Which realised the issue wasn’t that nobody wanted them but that the ones they could buy - that TR7 - weren’t very appealing.
Mazda spotted an opportunity and, calculating and minimising the risk, decided to give the world what it had already had: a Lotus Elan. The Elan was arguably ‘peak sports car’, a small, nimble, reasonably quick car that was easy to own. At least when it actually worked, which was less often than All The Time. Quite a lot less often.
Of course, Mazda didn’t simply rejuvenate the Elan. It did the next best thing, recycling the basic shape and replicating the small, practical, ‘just fast enough’ feel. And crucially ironing out the stuff about not working that often. This was a Mazda after all.
The MX5 was launched in 1989. It was a huge success, purely because Mazda gave the world what the world really wanted. Never ones to let an opportunity pass by, particularly one they hadn’t taken any risk creating, the world’s car makers followed Mazda’s lead and launched their own takes on the small convertible vibe.
Each car maker created a car that appealed to their market and, in so doing, moved things slightly away from the Mazda’s turf. They created a mini motoring revolution that lasted from the mid to late 90s.
Here are the cars that put sporting back into two seater motoring.
It all started here. Or rather, as we’ve established, about 25 years earlier with the Lotus Elan. The Mazda was the result of ‘blue sky thinking’ at the firm’s Californian design studio, but in reality there was nothing new about the MX5. Except that it was really, really good.
Where previous inexpensive sports cars came with obvious flaws - popular as the MGB was, with its humble engine and stolid handling, calling it a sports car was always a stretch - the MX5 had none of those. It was brilliant fun to drive. It was reliable. Comfortable. It had quite a decent boot. It was made by Mazda. And it had pop up headlights.
Mazda launched several versions of the MX5 between 1989 and the Noughties, all of which tended to evolve rather than rethink the idea, leading inevitably to the car feeling as anachronistic as those final MGBs in 1980. But then Mazda entirely rethought the concept and injected the MX5 with new verve. The current car looks ace, drives brilliantly and is once again highly desirable.
At about the same time as Mazda was tinkering with its Elan clone, Lotus was thinking along much the same lines: creating a small, inexpensive sports car to fill the gap left by the MGB.
The resulting M100 Elan, launched like the MX5 in 1989, didn’t quite deliver as intended. Firstly, it suffered from comparisons with the car it should have been - the MX5. Despite being arguably one of the finest handling cars of its generation, the M100 looked dumpy compared to the MX5. It was designed by Peter Stephens, who penned the McLaren F1. There is an apocryphal story that someone made a mistake when transcribing the dimensions into the production designs, resulting in a car that doesn't quite look right. It was also front wheel drive: as good as Lotus had made it, purists were appalled. And it was expensive.
It didn’t help that Lotus also made it. Because despite engines and engineering borrowed from Japanese firms, trusting the assembly of those promising components to a firm not strictly familiar with quality and care was a recipe for disaster. M100 buyers quickly morphed from enthusiasm to outright frustration with their recalcitrant cars.
The Elan cost a fortune to make so it was expensive to buy, which meant it completely missed its target of being an updated, inexpensive version of the original Elan.
Then there were those comparison with Mazda's would-be Elan. The Japanese car showed that updating an icon for modern drivers was entirely possible - and yet the company best placed to do it had clearly fumbled it.
The Elan soldiered on through a confusing variety of relaunches, none of which succeeded in re-igniting sales. Only around 5,200 were sold worldwide in 10 years. Mazda sold more MX5s than that in a month.
Here’s another car that wasn’t really spawned by the MX5, but deserves a part in this story, if only because it’s so often overlooked.
While Maxda’s designers were sipping their lattes in California and pondering the future of sports cars, over in less sunny and less latte-orientated Tamworth, in the West Midlands, a smaller team of probably less tanned designers was doing something similar.
Spurred on by the death of the MGB and the perennial popularity of its Princess Anne-bating sporting estate, Reliant decided to get into the small sports car game. In some ways this made sense, because Reliant made the cheap Rialto so it presumably knew a thing or two about engineering low cost cars for mass production.
The firm envisaged a cheap, fun and simple convertible that would update the MGB for a more modern generation. Sales targets were set at a modest 2,000 per year.
Reliant commissioned Michelotti, famed designer of numerous Triumph sports cars, to create its new model. Sadly, as the finished result shows, the man who designed the TR4 wasn’t quite at the top of his game in the early 80s when Reliant approached him.
The resulting glass fibre SS1, launched in 1984, looked quite a lot like traditionalists might want a sports car to look, but with some odd and rather ugly detailing. The interior was a sea of depressing grey plastic, with all the structural rigidity of a wiggly worm.
There were some quite good engines, including a turbocharged Nissan unit, and it was rear wheel drive, so it was fun in a very traditional sense.
The elephant in the room, however, was the firm that made it. Calling it a Scimitar didn’t fool many people that this was in fact the same outfit that gave the world the Reliant Rialto (nee Robin) that all the world hadn’t quite laughed at enough yet. Also, it was catastrophically badly built. For example, to reduce costs Reliant only cut out the small cooling ‘gills’ on one of the car’s front wings. To avoid the other side looking odd they simply stuck black masking tape over the uncut ‘gill’ to make it look like it had been cut out.
It turned out that selling an ugly sports car made by the same people who put wheels beneath Princess Anne and 25% fewer wheels under miners spending their redundancy cheques wasn’t the greatest draw for sports car buyers.
Reliant soldiered on with the SS1 until, remarkably, 1995, including a redesign - that made the car look half decent - and a rename, that removed the unpleasant Nazi associations. They sold 1,507 in total, quite a long way short of that annual target.
MGF enthusiasts will doubtless baulk at the idea that the MGF owes anything to the MX5, least of all its existence. But its hard to argue otherwise.
It is true that MG had been planning a MGB replacement almost since the launch of the original car in the early 1960s. That project eventually became the TR7 whose woeful sales seemed to convince Austin Rover to stop bothering with sports cars and build its bread and butter saloons instead. In 1991, perhaps - or perhaps not - influenced by the newly launched MX5, MG Rover began developing the car that would become the F. Various ideas were tried, including a retro update that combined ideas from the B and Austin Healey (that proposal is on display at the Gaydon British Motor Heritage museum) before the firm honed in on a radical mid engined car that had echoes of MGB styling but was very clearly a brand new idea.
The F managed, despite the obvious weight of expectation, to wow critics and car buyers when it launched in 1995. It outsold the MX5, no small achievement, in part by aiming for the same market but doing things differently.
MG extended the car’s life with an update in the late 90s and then significantly redesigned the car for the new Millennium to create the MG TF.
The F was never the last word in driver engagement and suffered from shoddy build and a propensity to eat head gaskets. But owners loved them, able to overlook its Metro underpinnings to experience a clever, distinctive and capable sports car that was arguably the best MG that MG ever made.
The F’s success led to MG Rover ‘MG-ifying’ its saloon car range and ensured the survival of the only remaining British Leyland brand into the modern day.
Having tried and failed to out-Elan the MX5, Lotus went back to the drawing board. The firm still wanted to fulfill demand for a low cost sports car but realised an entirely different approach was required.
That approach was the Elise. Coincidentally borrowing the layout and - less coincidentally - the engine from the MGF, the Elise was an extremely focussed sports car quite unlike the MG. Light - in true Lotus tradition - with few compromises made to comfort, luxury or anything not strictly essential to driving, the Elise was and is a revelation. Just as the MX5 delivered fully on its own brief, so did the Elise. It appealed to ‘proper’ drivers, being exciting and engaging in ways nothing without a Ferrari or Porsche badge had ever been. But unlike them it was available at a real world price. Lotus made the lack of compromise its selling point: if you wanted air con or comfy seats, you weren’t an Elise customer.
The Elise’s simplicity also played to Lotus’ strengths. Where the M100 had been complicated, the Elise’s absence of ‘stuff’ made it much easier to build reliably. There are issues with the Elise, but far fewer than any previous Lotus.
The Elise remained in production for 25 years, evolving into the more supercar-orientated Exige. In its entire history Lotus has built around 70,000 cars. Over 20,000 were Elises and Elise versions.
In the early 1990s BMW was riding high. Then got itself into a bit of a pickle. Observing VW hoovering up ailing marques like Seat and Skoda, then turning them around, and Mercedes dramatically pushing into a mass market future with the A Class (provided that future didn’t involve too many elks), BMW decided it needed to do two things to survive. Firstly, it needed to find its owning ailing car maker and quickly. And secondly it needed to go into the mass market. I think we can see where the roots of that strategy lay.
Proving the old adage that what works for someone else isn’t guaranteed to work for you, BMW’s acquisition of the remains of Austin Rover didn’t go entirely smoothly. It did sort of enable the firm to fast track itself into the SUV market thanks to Land Rover know how, and - eventually - push into the mass market with the MINI, but in all other respects it was a catastrophe.
Less painful was the firm’s other moves towards the mass market. Its first foray was the 3 Series Compact, which took a 3 Series saloon, shortened it and added a hatchback. The car was ok but it felt like BMW wasn’t really trying.
Its next attempt to broaden its market was altogether more successful. Taking a cue from the MX5‘s success, BMW decided that its sporting credentials ideally suited the market and it set about developing what became the Z3. Launched in 1995 and featuring a lot of proven 3 Series mechanicals, the rear drive Z3 looked the part - its shark-nosed styling was very ‘on trend’ in the retro-obsessed mid 90s car market - and drove well too.
The trouble was, in a drive to out MX5 the MX5, BMW had miscalculated. To keep prices attractive and simplify production, the Z3 was launched - like the MX5 - with just one engine. While this didn’t really matter to the Mazda’s buyers, who wanted a simple sports car, buyers expected more from BMW. And the 140 bhp 1.9 4 cylinder engine in the Z3 definitely didn’t cut the mustard.
Consequently, the Z3’s launch didn’t go as well as planned. The car earned a reputation for not really being in the spirit of sporting BMWs. This criticism tended to overlook what was great about the car - it was comfortable, it handled well and it was extremely well built. But it was quite expensive compared to the Mazda and the Japanese car was - whisper it - better to drive.
BMW quickly realised its mistake and dropped its grunty 2.8 litre straight six into the Z3, at the same time redesigning it with a wider track for improved looks and handling. More engine options followed.
These changes did the job. They moved the Z3 away from the MX5 and into a familiar market for BMW - premium sports cars. Shorn of this direct comparison the Z3 flowered into a distinctive, supremely capable sports car that was engaging and fun yet more mature and useable than the Mazda.
The Z3 may not have delivered on BMW’s plans to head mass market but with clever development it became a really good car.
We have an early 2.8 ‘widebody’ Z3 on our classic car hire fleet. It consistently receives positive reviews. We think of it as a sort of latterday Austin Healey - great style, grunty engine, upright driving position and comfortable too.
Audi was late to the 90s sports convertible pile-on, launching the TT roadster in 1999, but it had clearly spent that extra time perfecting things.
The TT, originally shown at the Frankfurt Motor Show in 1995, went from design proposal to finished car virtually unchanged. A clever rethink of the 1960s NSU TT shape, it was intended to stake Audi’s claim to the emerging coupe market of the 90s, which appeared to be replacing the hot hatch one of the 80s. Utilising Golf oily bits below the surface, the TT had 4WD and up to 225 bhp, making it quick and sure-footed to go with the beautiful exterior.
The MX5’s success made a convertible version inevitable and it appeared a year after the TT’s 1998 launch. Audi managed to cleverly morph the coupe profile into a soft top. But this was a car specced and priced to sit a long way above the MX5. Unlike BMW with the first Z3s, Audi really knew its market.
Despite the power and the benefit of all wheel drive, where the simpler, slower Mazda was a proper driver’s car, the Audi struggled to be really engaging. It was quick and went round corners as if it was superglued to them, but it wasn’t quite as fun and didn’t require quite as much skill to do it. Or, indeed, any skill.
The TT is a great car and the roadster a great place to be on a sunny day. Accept that the technology is doing most of the work and you have a clever, distinctive sports car that is guaranteed to get you to your destination. Whatever the weather does.
Honda is, by any reasonable measure, an odd car company. The maker of the excellent but uber conservative Accord has also branched into considerably different areas. And sometimes outright madness. Take, for example, the NSX, one of the greatest supercars. With the wrong badge. And recent Civics, which take whacky as a design principle.
In the mid 90s the firm decided to combine these disparate strands into one model, the unassumingly named S2000. On the outside the S2000 looked distinctly ordinary - pleasant, like an Accord.
Under the bonnet, however, things went a little, well, mad. The 2 litre engine, on the surface, seemed nothing special. Most cars in this class had similar size motors. For the madness you had to delve into the stats. And drive it. With up to 125 bhp/litre Honda claimed it had the highest specific output of any mass produced engine. In the world.
As road testers quickly discovered though, the S2000 wasn’t just about Top Trump claims. It delivered where it mattered, from behind the wheel. And it did that by revving to an astonishing 9,000 rpm. Most rivals provided peak power at around half that. Put very simply, the powerful, revvy S2000 was an absolute hoot.
Of course, because Honda had developed it, it handled brilliantly too, with excellent weight distribution and nimble, tactile handling.
Despite all this, plus the awards and plaudits that come with just being a really good car, the S2000 was never much of a sales success. Less than 130,000 were sold worldwide in 11 years and fewer than 20,000 found homes across Europe. This may have been because the S2000 never quite fitted a market segment. Too big and clever to be a MX5 rival, too Japanese to be accepted as a premium market player in Europe. It may also be true that the engine’s trickery really only motivated people who write about cars, plus a handful of enthusiasts.
Today the S2000 is largely overlooked beyond a small but admittedly growing band of enthusiasts. Arguably it is, aside from the MX5 and Elise, the best car on this list.
In the early 1990s the management at Mercedes had a revelation. Instead of engineering at great expense and building, similarly at great cost, stolid, safe cars that literally lasted beyond time, wouldn’t it be far easier to pretend to do both of those things?
Selling The People the idea of a Mercedes was surely far simpler, and much more profitable, than actually selling them a properly engineered and built Mercedes? This idea appealed in large part because those people going into Mercedes showrooms probably wouldn’t notice the conceit. That would be the job of buyers further along in the car’s life.
Alongside this discovery Mercedes also realised that to survive in a car market that was consolidating and becoming increasingly driven by high volume economics, they needed to make more cars. Quite a lot more cars. So the firm began to explore ways to put Mercedes cars on the driveways of people who may never have considered buying a Mercedes. And quite probably didn’t have a driveway.
The first fruit of this exercise was the A Class, a small MPV style hatchback that was radically different to anything Mercedes had sold before. It also provided the first indication that Mercedes might have dialled things back a little on the engineering front. Certainly elks, which the car apparently proved unable to avoid without toppling over, would surely agree if given the opportunity to comment.
Mercedes' next foray into a wider market was much more successful, in part because it involved doing what it already did well, but smaller. The SLK was to all intents and purposes a small version of the SL sports car. Which was handy because that is literally what its name told you it was.
To create the SLK, Mercedes simply shrunk the SL but retained everything that made that car so good - which it should have been since Mercedes had been making SLs for decades. In particular it got a version of the SL’s brilliant folding hard top roof, which turned a SLK coupe into a SLK convertible at the press of a switch.
The SLK was clever, therefore, but also good looking and good to drive. Although most were sold as autos, rear wheel drive and decent steering meant the SLK could be properly hustled like a MX5. Unlike a MX5 it was also relaxing to drive on motorways at high speed. Also unlike the MX5 there was a huge range of engine choices plus the option to personalise the interior.
The SLK was expensive though and, unfortunately, the product of Mercedes’ de-engineering process. The interiors proved less than hardwearing. The bodies rusted. Things, like the complicated roof tended to have a distinctly un-Mercedes like tendency to stop working.
All of which means you can buy an early SLK for much less than an equivalent MX5. And you should. Despite the inherent risks - fixing that roof can cost upwards of £3,000 - they are so cheap that it is a car to simply enjoy as long as it lasts.
Before Mazda stole the blueprints to the Lotus Elan and gave the world the MX5, Toyota was doing something quite similar with the also quite similarly named MR2. The original MR2 took as its template the brilliant Fiat X1/9, a mid-engined mini Ferrari that had sold by the bucketload (shortly thereafter to rust away and be quietly swept up in the same receptical).
Where the X1/9 was almostly comically fragile, the MR2 was designed and built by Toyota so it was brilliantly engineered and reliable. It was great to drive too.
You would imagine that being early to the sports car game, Toyota would have stolen a march on Mazda when it came to launching a replacement for the original car. Unfortunately it fumbled the ball. The second generation MR2, which looked remarkably like the still-born Panther Solo, was bigger and much more of a coupe with targa roof than a car that might steal the MX5’s cheap and cheerful thunder.
When it came to replacing this car Toyota clearly realised the error of its ways. The third generation MR2 was very clearly a MX5 rival: small, light, relatively low powered, a simple to grasp model range and a proper convertible. But still mid-engined, so a hoot to drive - and it really was, arguably surpassing the original, which has set benchmarks in its class.
This car was really very good. Entertaining, good value and built by Toyota. But it was a sales disaster, barely nudging the dial compared to the MX5. This seems to be because it revisited the original concept, rather than the Series 2, leaving buyers hoping to trade in their cars somewhat confused.
The MR2 also didn’t seem to look quite right. Where the MX5 - mainly by copying another car that did look right - did look good, the Toyota somehow looked like it was two front ends of the same car stuck together. Those looks have certainly mellowed, and it’s hardly unattractive, but at the time they contributed to the sense that Toyota had worked hard to out-MX5 the MX5 and just fallen short. Buyers presumably concluded that if you wanted a MX5, you bought a MX5.
Today MR2 values are on the floor. Buy a later car, on which Toyota fixed a potentially catastrophic engine problem, and you’re getting one of the finest handling cars of the 1990s for peanuts. Alfa Romeo Spider
Strictly speaking Alfa didn’t launch a MX5 rival because it never really stopped making a MGB rival. By the late 80s, when the MX5 hit showrooms, Alfa was still churning out the original 1960s Alfa Spider, albeit with some very 80s squishy spoilers bolted on as ‘improvements.’
It is possible though that the decision to roll the classic Spider dice once again was at least subconsciously linked to the rejuvenation of sports car motoring brought about by the MX5. The Series 4 version of 1990 was a comprehensive makeover of the nearly 30 year old design, integrating plastic bumpers into a revamped body and adding fuel injection to the venerable twin cam engine.
The Series 4, which was actually built by Pininfarina, not Alfa, did for Alfa what the MGRV8 did for MG: reminded sports car lovers that Alfa existed.
The Series 4 held Alfa’s place in the sales queue while it developed a wholly new car, the 916 Spider and coupe of 1995. Although clearly influenced more by its predecessors than anything from Japan, the new Spider arguably wouldn’t have got the green light at cash-strapped Alfa if the MX5 hadn’t reopened the market for drop top sports cars.
The old and new Spiders are very different cars - one is rear drive, one front for example - but they are distinctly Alfa. In the sense that many other cars in their sector do what they do much better, but both cars could only be Alfas. They are characterful and fun to drive and bestowed with brilliant engines.
At Great Driving Days we both versions of these cars on our classic car hire fleet.
In the 1990s Porsche was still grappling with an existential problem that had bedevilled it since the early 1970s. In order to grow and be sustainable the company needed to make more cars. To make more cars it needed more models. But each time it tried to launch a new model, like the 928, 924, 944 and even the 914, customers reminded it - mainly by not buying them - that they really only wanted a 911 from Porsche.
Before Porsche owners press infuriated digit to keypad, I realise that each of those cars was fairly successful. But they didn’t do what Porsche intended: create a 911 killer, a car that would burnish the firm’s reputation beyond making cars that went through hedges backwards.
That job fell to the Porsche Boxster, launched in 1996. On paper the Boxster looked likely to face a similar future to all those other non-911s. At launch, it was fairly low on power - about 200 bhp - and it was entering a market that really didn’t quite exist. It aimed to be a sort of MX5 for posh people.
That this market even existed on paper was because the MX5 had opened the door. Porsche realised that the love of going fast with the top down in a really, really good sports car wasn’t the preserve of those who bought Mazdas. Rich or poor, our visceral motivations are broadly similar.
The resulting mid-engined Boxster did a clever thing that secured its future. Apart from being astonishingly good to drive, it looked an awful lot like a contemporary 911. Where previous Not 911s had tried to furrow their own aesthetic path, the Boxster made no secret of its association. The whole front end of the first generation Boxster is identical to the 911, allowing Porsche to cut manufacturing costs and give the new car a ‘halo effect’ from the old one. A lot of the oily bits and most of the interior were also shared. With hindsight all this seems obvious, but it was a crucial part of the Boxster’s early success. Boxster buyers could easily fool other drivers that they were in fact 911 owners.
Had the Boxster just been a cheap 911 copy then of course it would have failed. But like the Mazda MX5 it was very, very good. Nimble, nicely balanced and just quick enough; it was also practical - with two large boots - comfortable and virtually bulletproof.
The Boxster’s success was the first step towards Porsche’s current sales hegemony. All subsequent Porsches have followed the 911 Lite design formula, to considerable success.
We have an early Boxster on our classic car hire fleet. And I can confirm it is utterly sublime.
The S was the first fruits of TVR’s new owner, Peter Wheeler. Bearing a striking similarity to the late 70s 3000, the S - launched in 1986 - was borne of a similar imperative to the Scimitar SS1 and MX5: to provide a relatively cheap and relatively simple sports car in the spirit of all those great British sports cars of the 1960s.