Dynamism and Slow Adaptation: The Software Industry in Santa Catarina / Brazil (1)

Jörg Meyer-Stamer
German Development Institute, Berlin

September 1996

Santa Catarina, one of the southern states of Brazil, is remarkable in several respects, one of them being a rapidly growing software industry. As the state is the home of a number of leading Brazilian industrial firms, it has a long story of using data processing technology. A few leading firms founded in-house data processing centers or separate DP bureaus in the mid-1960s. They have served as the nucleus from which numerous software firms have spun off since the 1980s, sensing that a window of opportunity had opened with the introduction of mini- and later microcomputers. This process gained in importance as DP bureaus were consistently losing business by the late 1980s because customers were switching to personal computers. The option was either exit or adaptation, i.e. turning themselves into software houses. Today Datasul, one of the local firms, is probably the largest Brazilian software firm; it sells integrated packages for the manufacturing industry (similar to SAP).

Software firms in Santa Catarina are mostly focusing on two types of products:

(1) packaged software for standard applications with a large number of customers which is distributed via shops (bookkeeping, word processing),

(2) integrated software for different types of customers, e.g. industrial firms, medical laboratories, or municipal administrations.

Economic performance of the firms

The overall economic performance of the industry is good. A number of firms have experienced dramatic growth processes over the last years. This has to do with the fact that Brazil's market for computer hardware and software has been growing rapidly, recovering from the era of the "informatics policy" which brought high-price, low-quality products. Accordingly, the economic climate in the software industry is much healthier than in other sectors of the Brazilian economy. This situation will probably prevail for some time as the degree of informatization in Brazil is comparatively low and there is a lot of pent-up demand. Consequently, the success of software firms depends on fundamental entrepreneurial qualities like a sense for business opportunities, a good marketing effort, and flexible adaptation to changes in demand.

In order to assess the competitive potential of software firms in Santa Catarina, it is useful to look at the four factors identified by Porter (1990) as key determinants of industrial success.

Business strategies and structures and rivalry: Porter notes that despite all differences and national peculiarities one characteristic shared by competitive economies is that there is sharp competition among national firms. Statistically, national champions may enjoy advantages of scale; but the real world is dominated by dynamic conditions, and here it is direct competition that impels firms to work for increases in productivity and innovations. In this context, anonymous competition often turns into concrete rivalries and feuds, in particular when competitors are spatially concentrated. "The more localized the rivalry, the more intense. And the more intense, the better." (Porter 1990, 83) This is all the more true, as its effect is to cancel out static locational advantages and compels firms to develop dynamic advantages.

The performance of software firms in Santa Catarina in terms of organization and strategy is mixed. Some of the leading firms have well-defined strategies and are impressively well organized. In fact, local rivalry is an important factor to explain this point; for instance, the four leading firms of integrated software for industrial firms are all located in the same city, Joinville. Likewise, the two leading producers of software packages are located in Blumenau. The leading firm in integrated software for industry is currently preparing for its ISO 9000 certification. On the other hand, several firms had serious deficiencies in management and, in particular, in marketing. We observed various cases of technology-driven firms with a nice product but without professional marketing efforts and, accordingly, with a below-average performance a typical feature of technology ventures yet a strategic error that is so obvious that one should expect firms to avoid it.

Factor conditions: These include, among others, the availability of qualified manpower and adequate infrastructure. "Contrary to conventional wisdom, simply having a general work force that is high school or even college educated represents no competitive advantage in modern international competition. To support competitive advantage, a factor must be highly specialized to an industry's particular needs a scientific institute specialized in optics, a pool of venture capital to fund software companies. These factors are more scarce, more difficult for foreign competitors to imitate and they require sustained investment to create." (Porter 1990, 78)

The situation in Santa Catarina is a kind of paradox. It soon becomes obvious to the visiting researcher that the institutions in the region are not in a position to provide local firms with specialized factors. However, firms are on the whole happy with what they obtain.

The supporting structure consists largely of five types of institutions: First, training institutions at secondary and tertiary level; second, technology incubators; third, a technology institute specialized in information technology; fourth, infrastructure providers; fifth, financing institutions.

As regards training, most firms recruit young people who have a secondary degree, i.e. the Brazilian equivalent of high school graduates. Firms are reluctant to recruit university graduates not so much because they are not available but rather because the firms feel that university education does not provide particularly valuable skills. Firms rather prefer to train young people in-house. The major complaint among firms is that they cannot recruit the kind of employees they would like to have. This, however, reflects an unrealistic hope, namely to find experienced programmers and systems analysts on the market, something that is unlikely in an environment of a rapidly growing industry. Moreover, firms hesitate to recruit employees from outside the region as they fear that they will too soon return to more glamorous places like São Paulo (which firms in Santa Catarina consider the software capital of Brazil).

Regarding technology incubators, it is important to point out that the software firms were there before the incubators. This does not imply a critique against the foundation of incubators. On the contrary, it reflects the fact that government has noticed the dynamism in this industry and is trying to stimulate it further. Software incubators, or technology incubators that also house software firms, so far exist in three cities (Blusoft/Blumenau, Softville/Joinville, and Softpolis and Celta in Florianópolis). The foundation of the incubators was in two cases plagued by delays which were mostly due to the fact that the government did not allocate the funds it had promised. In this respect, one cannot fail to notice the gap between the ballyhoo the state government makes about its effort to stimulate the software industry and the actual practice. Yet the performance of the incubators so far looks quite promising as they have a substantial number of applications of firms that want to enter, and so far hardly any business failures.

With respect to the information technology institute CTAI, which was founded in 1993, its mission does not yet seem entirely clear. The institute being located in Florianópolis, its networking with local firms or firms elsewhere in the state is as yet weak. It is trying to position itself as a center of excellence in multimedia development, and as a training center in different fields of information technology. Yet in contrast to other initiatives, firms did not refer to any experiences of cooperation with CTAI when we interviewed them. This need not be the sole responsibility of CTAI but it reflects the fact that it has not established itself as an institution that a serious software firm cannot ignore.

Regarding infrastructure, there a few serious bottlenecks. The state phone company has been upgrading its networking over the last years so that sufficient, and sufficiently fast, links are available. Education institutions are hooked onto the Internet with a 2 Megabit link. Occasionally these services are not available as service providers are still high on the learning curve.

Regarding financing institutions, most firms stated that it is complicated to get access to credit or capital. The establishment of a venture capital fund which was led by the federation of industries of the state reflects this problem as well as the determination of some actors to do something about it. The venture capital fund will largely cater to firms which have already established themselves on the market and need to finance an expansion process. At the same time, some firms have found creative ways of dealing with the financing problems, for instance by striking deals with private banks which are interested in benefiting from the dynamism of the industry and are not overly risk- averse, or by negotiating leasing deals with equipment suppliers.

Existence or lack of related and supporting industries: Spatial proximity of upstream or downstream industries facilitates the exchange of information and promotes a continuous exchange of ideas and innovations. Porter refers, among other things, to experiences with industrial districts in Italy. There, it is not only the interaction between firms along the value chain but also close contact and exchange between competitors ("the information is in the air") that establishes competitive advantages as it creates various externalities and stimulates cumulative learning processes.

In this respect, the situation in Santa Catarina is rather disappointing. Interaction between firms along the value chain (e.g. with respect to vendors of development tools and of final products) hardly exists. Even in terms of non-software inputs (e.g. editing and publishing of manuals, advertising) local software firms often rely on companies outside the region. The picture is somewhat different when it comes to contacts with customers. But even here those firms which have a close contact with customers, like the vendors of integrated software for industry, do not have exceptionally close links with local customers, at least not in the sense that close contact with local firms would create specific learning-by-interacting advantages.

The situation is even worse when it comes to contacts between rivals. It has often been noted that the kind of informal and formal cooperation between firms that is often observed in industrialized countries is a rare phenomenon in Brazil, and we found this to be a salient feature in most industrial branches we researched in Santa Catarina; these included the textiles and garments, ceramic tile, and electromechanical and metal engineering industry, see Meyer- Stamer (1996). Between software firms, there is often not so much as no cooperation but actually aversion, if not hatred. In some cases, this has can be explained with the fact that some firms have lured employees away from local rivals, or employees founded their own firms, taking proprietary know-how with them. In other cases, we found it complicated to come about with any straightforward explanation. More likely than not this behavior reflects the local business culture plus the fact that the firms have been doing fine so far without cooperation; the only case of close cooperation we found at all was in the ceramic tile industry where it emerged as a reaction to a deep crisis. Moreover, there is also a high degree of ignorance in each of the regions where we conducted interviews regarding the situation in other regions; the interviewed persons would believe that the situation of firms was more favorable in the other regions. The only exception regarding informal cooperation in the software industry are newly-founded firms in the incubators which can hardly avoid to get in contact with each other.

Demand conditions: The more demanding the consumers in an economy, the greater the pressure firms face to constantly improve their competitiveness via innovative products, through high quality, and so on.

It is here that software firms in Santa Catarina, like elsewhere in Brazil, probably face the most serious constraint to sustained competitiveness. Largely due to the informatics policy which unsuccessfully tried to stimulate the emergence of a national computer hardware industry (Meyer-Stamer 1992), users (in particular PC users) in Brazil until recently had to deal with overpriced and technology outdated products which were, as a rule, two years behind the international standard. The migration from DOS to Windows 3.1 therefore occurred much later in Brazil than elsewhere as machines were widely in use that did not run Windows with acceptable speed. Moreover, it was somewhat unpredictable when the migration was to take place (just like currently nobody knows how fast Windows 95 will spread in the country, and if and when the Internet will become important as a platform in Brazil). Therefore, software vendors only belatedly started to migrate from one platform to the next. In the case of one firm it meant that it dropped from the leading position in DOS word processing to an also-ran status in Windows word processing. In the case of a medium vendor of integrated software for industry it meant that it has had hardly any sales for about a year as it could no longer sell its DOS-based product and had its Windows-based product not yet available.

Support policies

Software firms are benefiting from several types of support. First, there is support from the federal government, mostly linked to the Softex-2000 program. This program was launched in 1992 with the aim of establishing Brazil as a major exporter of software; the aim was to obtain 1 billion $ in software exports in 2000. The basic idea was to achieve this goal by stimulating the establishment of new, world market-oriented software firm (where world market mostly meant US market; an "outpost" for Brazilian software firms was set up in Miami). One important instrument is the support for software incubators which have sprung up in various places of Brazil, often catalyzed by the Softex program. By 1995, there were 14 such Softex-related incubators, three of them in Santa Catarina. These incubators receive some financial support and equipment and software donations via the Softex program. Moreover, there are other benefits like grants for firms that hire professionals for specific projects, and grants for visits to expositions and congresses abroad and to pull foreign specialists to Brazil.

The problem with the Softex program is that it links two concepts, each one alone being sensible, but they do not make sense if their are bundled together, namely supporting emergent software firms and stimulating exports. There are two reasons why recently founded software firms will probably not enter export markets. First, it is complicated. As mentioned before, firms are often weak when it comes to management and marketing anyway, and it is hard to imagine that they can make it in a far-away market like the US where competition is cutthroat, and, moreover, people do not speak Portuguese. Second, there is the fact that the domestic market is reasonably dynamic so that there is no reason to go abroad in the first place as a firm can make a reasonable profit within the borders of the country. In fact, we found few firms that are doing, or considering, exports. There was the largest firm which has been selling its integrated software to a few US firms, and there were a few other firms that were exporting, or considering to export, to neighboring countries like Argentine and Chile. It is not surprising that it is well-established rather than recent firms which are trying to establish themselves on foreign markets, be it that they sense business opportunities, be it that they are pulled abroad by clients.

Second, there is some support from the state and municipal governments. Both give some, albeit unpredictable, support to software incubators. The state government supports the presentation of local firms on the largest Brazilian computer exposition, Fenasoft. Municipal governments are reducing the poll tax of 5 % for software firms to between 0 and 2 %.

Prospects for software firms in Santa Catarina

The perspectives of software firms in Santa Catarina depend, quite naturally, on the overall development of the software market in Brazil. Provided that the market continues to grow, which is likely, and continues to follow a pattern of belated technological evolution firms in Santa Catarina will probably show a good performance. Yet it is unlikely that Brazil will not be affected by the radical change that is imminent in the software industry (see Survey The Software Industry, The Economist, May 25th, 1996; "Inside Microsoft", Business Week, July 15, 1996). Provided that the transition from the Windows-platform to the Internet-platform actually materializes, firms in Santa Catarina will get into trouble. This has to do, first of all, with the problem of a comparatively unsophisticated demand in Brazil which does not force, and not even encourage, firms to keep abreast of technological developments abroad. In other words, those firms which behave according to the predominating incentives and do not develop a more strategic vision are certain to face a bleak future. Moreover, there is the problem that although the Internet is increasingly popular in Brazil, we did not observe a particularly well-developed Internet culture. Firms somehow deal with the Net, but do it in a slow and defensive way. There also is the problem that it is somewhat difficult to actually use the opportunities the Net offers. For instance, in Florianópolis I found it quite complicated to get access to more than about 15 newsgroups. This reflects the fact that neither universities nor other entities have enough disk space available to keep the 6000+ newsgroups that actually exist, or even the several hundred newsgroups that are important for computer professionals. Moreover, there is the language problem as a graduate from secondary education is unlikely to be fluent in English.

There a few immediate policy conclusions that emerge from this reasoning. Policy is faced with the usual problem of technology policy that it can try to persuade or convince firms to do certain things, like behaving more strategically, but that it cannot enforce this. It is important here to point back to the fact mentioned before that within the current market it is perfectly rational for firms to adapt a wait-and-see attitude regarding possible technological upheavals. Thus, instruments which are highly popular in Brazil, like fiscal incentives to reward innovative behavior, are unlikely to make much difference. Instead, software-industry related policies should take care to continuously upgrade the existing hardware, as for instance demands on hard disk space and velocity of machines and connections tend to explode with an increasing utilization of the Net. Another necessity is to continuously upgrade the curricula of training institutions, and to provide them with the necessary hard- and software and Net connections. Finally, there is an essential task for managers of institutions like software incubators to continuously expose their clientele to awareness-building measures, like seminars and demonstrations. The instruments to do this, like grants of the Softex program, do exist.

Footnotes

(1) The field research was conducted between February and April and in August 1996 as part of a joint research project between German Development Institute and the Instituto Euvaldo Lodi of the Federation of Industries of Santa Catarina, in particular in the person of Silene Seibel. We are indebted to the Associations of Commerce and Industry in Blumenau, Criciúma, and Joinville that did a tremendous job in helping us to organize and in accompanying the fieldwork. I am also indebted to executives of firms, supporting institutions, and government agencies for sharing a lot of their valuable time with us.

Bibliography

Meyer-Stamer, Jörg, Filling the Local Space: Obstacles to Strenghthening Industrial Competitiveness on the Local and Regional Level - The Case of Santa Catarina / Brazil. Paper prepared for EADI Conference, Vienna, Sept. 11-14 (URL: http://www.cs.tu-berlin.de/~jms/sc-eadi.html), 1996

, The End of Brazil's Informatics Policy, in: Science and Public Policy, Vol.19, 1992, No.2

Porter, Michael. E., 1990, The Competitive Advantage of Nations, New York: The Free Press