Trojan Mule?

Was the Alfa Romeo Arna one of the ‘worst cars ever’?

Hybrid, of sorts: the 1983 Alfa Romeo Arna

At the beginning of the 1980’s Alfa Romeo was in grave trouble. Its reputation had been marred by the problems that afflicted its C-segment Alfasud. Built at the behest of the Italian government in a new factory in Pomigliano d’Arco near Naples, it was riddled with faults, the most serious of which was its tendency to dissolve into ferrous oxide at an alarming rate.

By 1980, the Alfasud’s build quality had improved noticeably, but not so Alfa Romeo’s reputation. A replacement model, the 33, was in development and would be launched in 1983. The 33 would be a somewhat larger and more expensive car, growing by just 20mm (¾”) in wheelbase, but by a more substantial 185mm (7¼”) in overall length. This left room for a smaller and cheaper car to replace the entry-level Alfasud. It is a moot point as to whether or not Alfa Romeo actually needed such a car in its range, unless it was really determined to challenge Fiat on the latter’s home ground, surely a foolhardy move? In any event, Alfa Romeo had neither the time nor resources to develop another new model.

Step forward Japanese automaker Nissan as a potential white knight. Nissan’s motives were, of course, not at all altruistic. In the early 1980’s there was growing hostility amongst Europe’s automakers concerning the invasion of Japanese cars that were eating into their traditional market shares. This was happening because the Japanese offered cars that, if dynamically underwhelming, were well built, reliable and generously specified. Rather than accept this and try to raise their game to match, the Europeans’ first instinct was to cry foul and call for import quotas and even an outright ban.

Nissan saw the opportunity to work with Alfa Romeo as a way of circumventing import quotas. The Italian government, weary of pumping money into state-owned and perennially loss-making Alfa Romeo, approved the deal despite opposition from politicians, unions and other manufacturers. Some commentators described it, and the contemporary Honda Ballade-based Triumph Acclaim, as Trojan horses for the Japanese auto industry’s wholesale invasion of Europe.

A joint-venture company, Alfa Romeo Nissan Autoveicoli S.p.A., was formed following the signing of a memorandum of understanding in October 1980 between Alfa Romeo and Nissan. For the venture, a new factory would be built at Pratola Serra near Naples.(1)

The outline of the new model was quickly agreed: it would comprise the body, interior and independent rear suspension and (drum) brakes from the forthcoming N12 generation Nissan Pulsar/Cherry and the engine, gearbox, steering, brakes and front suspension from the Alfasud. It was hoped that such a combination would produce a car with the character and brio of an Alfa Romeo, coupled with the build quality and reliability of a Nissan.

Weighed against these hopes probably should have been concerns about how the market might regard such a car. Given that it would look almost identical to the Pulsar/Cherry, could it ever be accepted as a proper Alfa? Moreover, the cost of re-engineering of the Nissan bodyshell to accommodate a flat-four longitudinal engine in place of a transverse inline unit would not be insignificant.

Christened the Arna,(2) it was launched at the Frankfurt motor show in September 1983. Like its progenitor, it was a not unpleasant if rather bland three and five-door B-segment hatchback. Visually, it differed little from the Nissan. At the front, an Alfa Romeo shield grille had been rather incongruously plonked in the centre of what looked like Nissan’s black plastic slatted affair (but wasn’t).

The differences were more marked at the rear: instead of the Nissan’s two-tier rear lights, wider and shallower units that bookended the number plate were fitted. This necessitated a rather makeshift looking black filler strip to be inserted between the lights and bumper as there were no metalwork changes to the rear panel. The Arna’s steel wheels with black centre covers were a recognisable carry-over from late model Alfasuds. Inside, it was pure Nissan, even down to the unusual oval instruments(3), although the steering wheel had been changed for a sporty three-spoke item with perforated metal spokes.

1985 Alfa Romeo Arna 1.5Ti interior. Image:

Under the bonnet, the Alfasud flat-four engine and transaxle gearbox were fitted. The former came in 1,186cc 62bhp (46kW), 1,351cc 70bhp (52kW) and later, 1,490cc 94bhp (70kW) forms while the latter was a five-speed manual unit. In a further nod to the ‘Sud, a second bulkhead had been inserted between the suspension turrets.

Autocar magazine road-tested the Arna 1.5Ti in July 1985. Its performance was “hardly in the really hot-hatchback bracket, but respectable” with a 0 to 60mph (97km/h) time of 10.23 seconds and a top speed (in fourth gear at the red line of 6,250rpm) of 109mph (176km/h). Oddly for a sporting hatchback, top gear was more of an economy overdrive, with a top speed of 104mph (168km/h) at only 4,750rpm.

Overall fuel consumption on test was surprisingly heavy at 24.3mpg (11.62 L/100km) but this was attributed to “the relatively harsh treatment the car received [on test] and owners should expect to achieve “nearer the gentle driving figure of 31.6mpg (8.94 L/100km) .” Steering was “a little heavy at low speeds but once above 40mph (65km/h) it is light and responsive.”  The predominant handling characteristic was “mild understeer” and lifting off abruptly in a bend just caused “the front end to tuck in controllably and scrub off excess speed.”

Despite the adoption of Nissan’s rear drum brakes in place of the Alfasud’s discs, braking performance was strong and fade-free. At lower speeds, the predominant noise is from the boxer engine, a sound “you either love or hate.” At higher speeds, particularly above 70mph (113km/h) wind noise is noticeable, albeit “not too overbearing.”

1983 Alfa Romeo Arna engine bay. Image:

Inside, the Nissan heating and ventilation system was “very effective” with an “excellent” supply of fresh air and a fan that was inaudible on its lower two settings. The Alfa Romeo-sourced seats were “firm with good lumbar and side support” but taller drivers found them “rather uncomfortable” and, perhaps surprisingly, the driving position was described as “typically Italian.” The front compartment was “very spacious with a pleasant feeling of airiness and plenty of headroom” while the rear was “a little cramped.” All-round visibility was “good.” Minor controls all worked well, and headlamps were effective on both main and dipped beams.

In summary the Arna 1.5Ti “stood up to our test very well” and “is very competitively priced at £5,590 (US $7,830).” The reviewer did, however wonder who might buy it: non-performance orientated drivers “will probably stick with the Cherry” and it’s a moot point as to whether “a ‘Sud driver [would] even consider buying an Arna.”

Autocar’s concerns about the perception and potential market for the Arna proved to be well founded. Production of ended in 1987 after just four years, with a total of 53,047 cars manufactured. The decision to cease production was the first taken by Fiat as Alfa Romeo’s new owner, which immediately terminated the partnership with Nissan.

1983 Alfa Romeo Arna. Image:

At this point, you might be wondering how the Arna, which seemed to perform perfectly well on test, ended up a much derided and even reviled car, and a fixture on those lazy and tedious worst cars of all time lists? The answer to this conundrum is, I believe, only in small part that its reputation was sullied by typical Italian build quality woes.

It is true that Nissan-branded versions of the Arna, sold under the name Cherry Europe, were nowhere near as well put together as the genuine Japanese-built Cherry(4) and often needed significant rectification work at Nissan dealerships before delivery to customers. Moreover, while the Arna was a perfectly acceptable car in itself, it was a poor facsimile of an Alfa Romeo, with its dowdy looks and lack of a pure-bred bloodline.

However, the main reason for the Arna’s terrible reputation is not that it was by any means a terrible car, but because there was such a degree of blind prejudice levelled against it. A large section of the automotive (and journalistic) community was just determined to hate it from the off.

(1) It appears that Alfa Romeo had learnt nothing from its gruelling experience with the Alfasud at Pomigliano d’Arco and was content to take the same risk again.

(2) This was simply an acronym of the joint-venture company’s name.

(3) A rev-counter was only fitted to the top of the range 1.5Ti model.

(4) At home, Nissan tried to capitalise on its Italian connection and sold a differently trimmed version of the JDM Pulsar/Cherry badged Nissan Pulsar Milano.

Author’s note: In using the word Mule in the title, I intended no slight against the Arna. The mule is, of course, a hard-working and dependable beast of burden.

Author: Daniel O'Callaghan

Shut-line obsessive...Hates rudeness, loves biscuits.

36 thoughts on “Trojan Mule?”

  1. Good morning, Daniel. Thanks for the excellent article about a car that I’ve only seen on two or three occasions. I knew very little about it, other than it was a Cherry with an Alfasud engine. Autocar’s conclusion was spot on. The Arna appealed to a very small number of buyers.

    I’m still thinking about the Alfasud mentioned here the other day. I wonder how long it will take for the Arna to escape my thoughts.

  2. Thanks for the reminder of this fascinating anomaly Daniel. Unlike Freerk, I don’t think I’ve ever actually seen an Arna in the metal, though I saw plenty of the Alfasud and 33 in my childhood (courtesy of my Alfa-loving godmother), and was unaware of its existence until fairly recently.

    When I first heard of the car, I wondered what Alfa and Nissan’s management teams had been drinking but, upon reflection, I think I can understand why it seemed like a good idea. If I’d been of car-buying age and means at the time, a car that went and steered like an Alfa with something of the reliability and quality of a Nissan might well have appealed. The Arna clearly fell short of that in reality but conceptually you can see what they were getting at.

    1. You just described what Alfa and Nissan management thought they would create: a car with Alfa flair and Nissan reliability and build quality.
      Regrettably it came the other way round.
      The Arna did nothing better (and many things worse) than the ‘Sud and it was just as bad where improvement would have mattered most.
      Why on earth after Alfa’s desastrous experiences with Pomigliano they had to build yet another factory near Naples is beyond my recognition.
      At least Pratola Serra is the main engine plant for FCA nowadays.

  3. Thank you Daniel, for an insightful article about that lost “gem” of Italian chaotic planning.
    Indeed the ARNA was a huge misstep, but leasons were learned, particularly with the new Pratola Sera plant. Nissan showed ALFA their expertize in automated manufacturing and quality control (which in turn was somewhat diminished by Italian laissez-faire practices), while on the other hand capitalizing in the Italian market as mentioned above. These techniques, along with significant adoption of automation and robotic assembly was implemented on the last model developed before FIAT’s takeover, the 164. The 90 successor proved to be a huge quality leap forward, even compared to the later, FIAT blessed (and engineered) 145/6. The 33, while better built than the Sud was a failure dynamically, completely missing the mark on handling, lightness and overall steering feel/response.
    All in all, the ARNA was a miscalculated experiment, mostly created to gain Nissan’s manufacturing techniques, while opening the EU’s flood gates for the completion of the “Japanese invasion”.
    The last remaining ARNA in Greece was purchased from a well-known collector a few weeks ago, heading for a ground-up restoration.

    1. The survival rate in Britain isn’t much better, if How Many Left is to be believed: there are three on the DVLA’s books, none of them licensed for the road.

    2. Please don´t mention the Alfa Romeo 90. I had safely forgotten my odd wish to drive or own one and now I have to battle this again. It´s quite enough to want a Trevi; adding the 90 to the yearning is a lot to deal with. Thanks.

  4. The tie-up was good in theory yet ultimately ended up being badly executed even if the Arna was not that bad a car, it was never really going to compare as a successor to the Alfasud visually.

    Not sure if the following Alfa Romeo Arna sketches would have been able to help salvage things.

    Also found this image of a 1985 Alfa Romeo Arna Pickup Prototype

    1. An Arna pick-up? What were they thinking?

      There wasn’t even a handy Datsun equivalent. The previous two Cherry derivatives had wagon and panel van derivatives with a beam rear axle and longitudinal leaf springs, but the N12 series comprised only the three and five door hatches and a saloon – and the odd EXA coupe.

      The putative pick-up is a pleasing-looking little thing, but doesn’t look terribly useful with that short load bed. Did Alfa see a market amongst Southern Italian smallholders, allied perhaps with their counterparts on the other side of the Ionian Sea?

      Perhaps Alfa felt that their commercial vehicle range needed to be expanded downwards, given that around the same time as the Arna misadventure they were selling the AR6 (Fiat Ducato) and AR8 (Fiat/IVECO Daily).

    2. Bob: thanks for those image. If and only if the Arna had been exactly like those drawings it might have worked as an Alfa. But like the Tagora drawings we showed here ages ago, the slightest deviation from those very wan forms led to a different car resulting. Those 80s rectilinear styles were very fragile meaning the drawings were hard to get into 3D without unwanted changes creeping in.

  5. Good morning all. Thanks for posting those images, Bob. Those unique slim rear lights are common to all Arna permutations. I find it odd that Alfa made more effort to make the tail of the car distinctive than they did with the front end. Perhaps they thought the Alfa shield grille was sufficient in this regard?

    Even if the Arna was a commercial failure, Alfa learnt something about automation from the project, as vkarikas has mentioned above, so it wasn’t totally in vain. I can count on one hand the number of Arnas I have seen over the years. Does that make the survivors desirable and collectable? I’m not sure.

    1. For rare cars like this prices can be anomalous. Someone asking 6000 euros for an Arna is aburd, even if it´s museum quality. At the tame very interesting and worthwhile cars like the Trevi and even Dedra languish in the doldrums. XMs are still absurdly cheap and so are Kappas. An Arna isn´t worth a hubcab of any of them.

    2. The problem with buying any ‘museum quality’ car for a substantial premium over a well cared for but regularly driven example is that, unless you continue to keep it pickled in aspic, it will immediately start to shed value.

      If I needed such a car, I’d be tempted by a good second-hand Giulia if I had access to an independent Alfa Romeo specialist. It’s the first Alfa for years that was designed from scratch as such (and may yet be the last, saloon at least).

  6. I don’t see anything inherently wrong with the JV, they were just wrong on the product. Going all in ro the C-segment where the VW Golf named the entire segment because of it’s proficiency was bound to be a disaster no matter how good the product.

    Instead they should’ve focused on what in the Japanese portfolio could compliment the Alfa line up and what could be a big niche seller in Europe. I can think of several things like the Hardbody truck and the Terrano SUV. Bertone hade done something similar with their Freeclimber that was a CKD Daihatsu Rocky mounted to a BMW six cylinder turbodiesel.

    Among the more interesting things they had was the Silvia sports car. They could’ve easily assembled the S12 and S13 Silvia as a replacement for the aging Alfetta GTV. The platform was rear wheel drive and fairly light going under 1200 kg. I think it would’ve been a success with the twin spark twin cam and the Busso V6.

    Why not a Laurel, Skyline, Cedric based car as a replacement for the Alfa Sei? Why not import something really small like a kei car or at least smaller than the Cherry? The sky’s the limit in this little thought experiment.

    1. The cars you imagine would have been yet more Alfas for people who wouldn’t buy an Alfa in the first place. There have been more than enough of those in the Alfa product history.

    2. I don’t understand what kind of people you’re referring to because the demographic who would buy an Alfa didn’t buy them anyway. Except for the latest Giulia platform Alfa Romeo has survived on shared Fiat-platforms for the last thirty five years, so I really don’t see a difference.

    3. Alfa Romeo South Africa built Daihatsu Charades in significant numbers at their Brits, Guateng factory from 1983-85. That was around the same time as the Arna, and with no re-badging pretences, nor Alfa content.

      That sort of thing might have worked in Italy, with more success than the Arna.

      Perhaps Alfa had higher ambitions than something akin to their 1960s arrangement with Renault. In 1964 Portello turned out 20,000 Dauphines – overall Alfa production in Italy was 87,000 vehicles. However bad or otherwise the Arna was, I find it astonishing that it was outsold eight years before by the ageing and not particularly well-regarded Dauphine.

    4. Alfa sold more than half a million 156s because they were like a gazelle and looked like it, had engines that either were as good as the ‘real stuff’ (Twin Spark) or the real stuff itself, were agile and light on their feet and ignored lousy crash safety and quality glitches because it was fun to drive. The 159 didn’t sell because it was heavy as lead, felt and looked like it and was anything but fun on the road. The Giulia doesn’t sell because the dealer network is crap.
      That’s nothing to do with demographic but with character of the car and the fun you get out of driving it.

    5. The Nissan 180SX was a very hot car in 1989. It sold fairly well on its 1.8 litre turbo mill. At the same time Alfa didn’t have anything but its aging 105 series Spider, and the GTV had already been out of production for a couple of years. I think they would’ve had a solid business case with that one.

    6. While understanding that the popularity of sportscars was waning during that period against the hot hatch prior to the Mazda MX-5.

      Always felt Nissan out of all the Japanese marques (and especially considering its historical ties to Austin) could have easily developed a suitable small front engined rear-wheel-drive 2-seater roadster during the 1970s-1990s+ (essentially a better Japanese-built Spridget) derived from a shortened Nissan S-platform (e.g. Silvia) to take on the Mazda MX-5 (similar to how the RX-7 FC also served as the basis for the MX-5 NA platform), the 1.8 Nissan CA18ET engined Reliant Scimitar SS1/SST/Sabre gives a rough idea of what am thinking (together with a better conceived Nissan EXA that was instead a RWD roadster/coupe).

      While Alfa Romeo had the aging Spider, the 1.4-1.7 Alfasud Flat-Four would have been a decent choice to power a (pre-BRZ) Alfa version of a hypothetical Nissan-derived MX-5 type roadster.

  7. I was curious to see just how badly the Giulia was selling, so here are its annual European sales numbers since launch:

    2020 7,436
    2019 10,932
    2018 17,075
    2017 24,679
    2016 10,475

    The situation in 2021 is even more grim: sales in the eight months to end August were just 4,008 units, which is 6,012 annualised.

    The Stelvio isn’t exactly setting the European market alight either:

    2020 17,438
    2019 26,866
    2018 30,099
    2017 17,159

    Sales in the eight months to end August were 10,532 units, which is 15,798 annualised.

    1. That’s when you try to sell a honestly good product through a dysfunctional dealer network specialised on delivering a spexial kind of non-service.

    2. I wonder if there are any decent independent Alfa Romeo specialists around who could look after a Giulia properly? If you had one locally, you could pick up a nice one-owner three-year-old 36k miles example for around £18k and it would be a nice left-field alternative to the default German trio:

    3. Daniel: In the South East of the UK there are at least three I can think right off the top of my head of who are longstanding, reputable and cater to Alfas of all ages. There are probably more. In fact, my understanding is that if you avoid the main dealers, you’ll probably be okay with a decent specialist. A place that caters to the classic side of the fence is usually a good omen in my experience…

    4. Freelance specialists certainly are a valid alternative for private owners where leasing contracts including service arrangements and extended warranty conditoins are not that important.
      In a market segment dominated by company cars you need official dealers that do their job properly because otherwise there won’t be Alfas to buy used.
      Alfa has a system they call ‘action cards’. An action card is a kind of voucher entitling the car’s owner to the work indicated on the card and the official dealer uses the card to get a refund from Alfa for parts and work. This system is open only to official dealers.
      They use this to roll out retrofit modifications to existing cars – not that the cars get necessarily better because the less the dealers tinker around with the cars the less trouble the owner has.

  8. One out of three ain’t bad, I’ve never heard a bad word from anyone who owned a NUMMI Chevy Nova/Geo Prizm/Pontiac Vibe or a Triumph Acclaim/Rover 200 mkI.

    IMO Marchionne-era FCA made two big mistakes; pouring resources into rehabilitating *both* Alfa Romeo and Maserati rather than picking one and expanding its’ line to cover both’s market space (while seemingly allowing the rest of the company apart from cash cows Jeep and Ram to wilt on the vine) and opening a separate Fiat-Alfa dealer network rather than selling them through the existing one which had finally been winnowed from three sales channels to one.
    Still better than Stellantis in Europe which now has, what, six dealer channels? Fiat, Alfa, Maserati (please tell me it’s usually either Fiat-Alfa or Alfa-Maserati even in Italy); Peugeot, Citroen-DS (again hopefully Citroen and DS is always a combined franchise even in France) and Opel-Vauxhall (at least no overlap here, Vauxhall in UK and Opel everywhere else). And however the American brands, especially Jeep, are sold there.

    1. At least in Germany the dealers are split Fiat-Alfa-Jeep and Ferrari-Maserati. Up to the era of the 159 Alfas were at least presented in separate showrooms but now it’s all merged into one and you have to find your way to the single Giulia on display through piles of American SUV bricks.
      What FCA really should have done was to invest into their dealer network to bring that forward to the standard set by the competitors in the sector. This probably would have cost several times the sum for the Giorgio platform and they simply didn’t have that money.
      The quality of the French side of Stellantis dealers isn’t really any better and they also don’t have the money needed to make any real changes.

  9. The following words are not mine, being a contribution to a publication for which I am responsible – but apposite here:

    “In the 1980s I was on a business trip to Pisa when I received a call to urgently visit another customer to finalise a contract.  This required a long fast evening drive to Vergeze in southern France. But the only car that the hire people would allow me to take out of Italy and leave in France was a rather ugly little hatchback.
    Driving it out of the hire depot I was immediately struck but the sheer verve and lovely free revving sound – so I set to and thoroughly enjoyed thrashing it, as one does small unwanted uncleaned Italian hire cars, along the Autoroute du Sud to my next morning’s destination.  Gave it an Italian service, as my sons would say.
    The next morning, a nice big order for our UK factory in my briefcase, I thrashed it even more enthusiastically to Montpellier Airport for a return flight home.  But before abandoning it in foreign parts I just had to peer under the bonnet. And there simmering under the oil mist and heat haze was an Alfa flat four.
    It turned out to be an Alfa Arna. Having a body designed by Datsun in Japan, and then built in an Italian factory, it was obviously doomed to be an ugly rust bucket and hence a car that Hertz Italy didn’t mind seeing the back of. Or perhaps that’s why they ended up as hire cars as no-one wanted to buy them! But what an engine!”

    1. That’s a great quote and probably sums up what Alfa intended: a more modern car than the ‘Sud, but with the same verve and (largely) road holding. A manufacturer has to sell a story as much as a product and in this case that story is just too complicated for the general public. In the case of the Arna: “it looks (and is meant to be built) like a Nissan, but underneath (especially engine and handling) there’s a lot of Alfa!” More or less the antithesis of badge engineering which – although it has a decidedly checkered history of its own, BL and GM spring to mind – seems to go over much better. Look at the VAG group, or the Toyota/Peugeot/Citroën AygoC107 triplets. Actually, in the latter case the French versions probably benefitted from the exact opposite story: “it looks like a French car with all the verve that implies, but it has Toyota quality underneath!”

  10. I think the ARNA was a product of pure desperation on Alfa‘s side and of „hell, why not“ attitude on Nissan‘s behalf.

    Alfa was ready to file for bankruptcy at the time the ARNA was conceived and had been for years already. They probably did not pay a single lire for the whole venture, poaching money-laden Japanese with the carrot of opening up the Italian and European markets.

    One must not underestimate the fear in Japan at the time of import restrictions. This is what lead them to show any interest in the train-wreck that BL was, what had them form alliances with GM (Nummi) or Chrysler (Mitsubishi) and then some. It is obvious there was no actual business logic beyond market entry for all of these ventures. Which is proven by the fact that all Japanese car-makers where happily making it on their own, as they simply started production on the premises (Sunderland, Swindon, all those factories in the US, etc.).

  11. I read with interest on ANE that Jean-Philippe Imparato plans to introduce an even smaller Alfa suv than the forthcoming Toenail.

    Which sounds like Arna thinking, but Mr. Learned is a canny fellow, and has stated that the way to success and profitability is to offer products in the market sectors where most customers make their purchases. Worked for Peugeot under his stewardship.

  12. I find the reviewer’s comment regarding the driving position rather suspicious. One wonders if he actually sat in it.

    1. Hi Simon. Is it the “typically Italian” comment you take issue with? I reported it faithfully, but did wonder how Alfa Romeo managed to achieve that in a Japanese car!

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: