Main page


by Valentino Piana (2014)




1. Significance

2. A formal and quantitative approach to relationships between competitors

3. An application to real-world data

  4. Dynamic and qualitative approaches to relationships among competitors  
  5. A few consulting venues from these analyses  


Who sells your product to your customer is your competitor. Competition is a triangular relationship involving two economic agents (typically firms) selling to a third one. We define competitors as "sellers of the same good to the same consumer". The latter is not only aware of the existence of the supply available but has a direct experience with both products, has found good arguments in favour of both, and is sensitive to further improvements in supply conditions (e.g. promotions).

We can relax these conditions in several directions, thus enlarging the definition and increasing the number of competitors:

1. competitors considered as "firms selling product similar to yours to customers similars to yours";

2. competitors as "firms selling a product which substitutes yours, by fulfilling the same need(s) but being largely different from yours in design and materials used";

3. competitors as "all firms having sold in the past your product to your customer".

As you can imagine, there are further mixes of conditions, including distinctions between producers, distributors and resellers, but you need to keep into account that the customer has a short list of products among which she or he actually makes an effort of comparison and choice, because of cognitive, emotional, time and attention limitations. Advertising is a way companies use to attempt to be included in this short list, even when their products have not yet been tested by consumers.

In this paper, we shall build upon our stricter and sharper definition of competitor to propose an exhaustive taxonomy of relationships among couples of competitors. We shall apply this taxonomy to real data, giving you the computer code to replicate our analysis to further data, including a database of more than 700 products. We shall discuss at length what characterises a "strong" competitor and what a "weak" competitor. We shall present a quantitative index of competitive strength.

Then we shall tackle the issue in a more dynamic and qualitative orientation, to include elements that our data could not cover, and share with you a few computer simulations that address, at least in part, these new features (e.g. product and process innovation, finance, advertising, etc).

In this context, several consulting venues are finally highlighted to continue our conversation.

2. A formal and quantitative approach to relationships between competitors

A firm selling a product in Argentina is not a competitor of a firm selling the same product in India: if the first succeeds, nothing can be said of the latter (success or failure is totally independent). Conversely, it is a competitor of another firm selling that product in Argentina as well (in the same cities and population segments): common demand trends involve both and, within the trends, the more the latter sells, the less the former. So they study each other's moves, promotions, introduction of new product variants, advertising and distribution networks. They imitate successful approaches to market, to the extent they are able and willing to do so. They cooperate in trade associations to lobby legislation and manage industrial relations with the workforce, retaining more or less degree of independence in setting wages and work conditions (be it in production or in distribution).

So the commonality of customers is a fundamental issue for competitors. Their incomes, tastes, decision-making rules, constraints, localisation, media exposure and absorbtion, social networks are relevant for all competitors and are subject to tracking, with some competitors possibly being better at it and building a competitive edge by deeper knowledge of demand. Conversely, a deeper knowledge of Indian customer would be of no use for the firm selling in Argentina.

In more general and formal terms, we define competitors as firms selling the same product to the same customer over a reasonable span of time (e.g. 1 year). We concentrate on couples of competitors, to open a venue for further structures when competitors are many.

For each customer, two firms can be competitors or not. So the number of customers for which they are competitors is a key statistics to measure the width of exposure to competition from each other.

The other key dimension is the respective strength of each competitor to the other. Later on, we shall discuss this strength in dynamic and qualitative terms, but for this section we use a simple statistics (sales) and two accompaning key data (the quantity sold and the price set or negotiated). Sales values (or "turnover") are equal to the price times the quantity.

In this vein, we consider "strong" the competitor that sells more than the other, which is then necessarily "weak". To be more precise, for each couple of competitors:

1. we call "strong" without further adjectives and nouns, the seller that sells more in both quantity and sales values, as well as charging a higher price;

2. we call "strong price undercutter" the seller that, by setting a lower price, sells larger quantities and obtain a higher turnover;

3. we call "strong premium niche" the seller that, by setting a higher price, sells lower quantities but still obtain a higher turnover;

4. we call "weak price undercutter" the seller that by setting a lower price, sells larger quantities but remains with a lower higher turnover than the other;

5. we call "weak premium niche" the seller that, by setting a higher price, sells lower quantities and remains with a lower turnover;

6. we call "weak" without further adjectives and nouns, the seller that, notwithstanding a lower price, sells less in both quantity and sales values.

In other words we highlight six roles based on strategy and success: the price competitor (lower prices, higher quantity) - which can be more or less successful (higher or lower values of sales); the premium niche (higher prices, lower quantities - symmetrically less or more successful than the price competitor); the weak (lower prices, low sales), who could also be considered as "fringe" or "marginal", and finally the strong (higher prices, quantities and sales).

The really "strong" competitor is capable of forcing a higher price and still selling more physical units of the good, so making a lot more in total sales.

These categories are mutually exclusive and exhaustive, except for the cases that see competitors having equal price and/or equal quantities and/or equal sales. You would need 13 categories to make the taxonomy including those cases, because many of the theoretical 27 possibilities are impossible, as values are not independent from prices and quantitites. For simplicity's sake, we use the reduced 6-roles taxonomy.

These six roles contain automatic symmetries, so there are just three types of couples:

1. one is "strong" and the other is "weak";

2. one is a strong premium niche and the other is a weak price undercutter,

3. one is a strong price undercutter and the other is a weak premium niche.

In other words, to charge a higher (premium) price can be effective, partially effective or ineffective to win over the competitor, respectively. Conversely, the three cases can be seen also as situations in which price undercutting is ineffective, partially effective or effective.

3. An application to real-world data

To get data at the level of single customer and have a list of his or her purchases over time in terms of single brands and product items across all the attended point of sales is fairly difficult. We present instead a large database of international trade where all countries in the world are included and, for more than 700 products, we have the quantity in weight, the unit price per kg and the total value of sales of each competitor in each market. It's a large subset of total world trade in year 2000, with more than 500 000 rows, in which you will find the product code (according to the 4-digit SITC international classification of goods), the importer, the exporter, the value sold, the quantity and the price [1].

Covered are the trade flows among 181 countries pairwise. We chose a mid-range product, with 56 exporters, 32 importers and a total of 226 flows (SITC: 3224 - Peat). This keeps the computational time in MS Excel in reasonable ranges. Meanwhile, for your experimentation we freely distribute the whole file and the computer code necessary to replicate our analysis to other markets (or to your own data). The code is in MS Visual Basic for Applications, embedded in an Excel file.

For this product, most countries do not have access to the good, since they do not import it, nor export (which can be taken as proof of presence of a domestic industry, normally selling also internally): consumers in only 66 countries can get the good (net of undocumented domestic industry which neither exports nor imports the good).

In a perfectly globalised world, all exporters should be present in all markets, so that all consumers face the same width of choice. By contrast, the maximum number of markets in which an exporter is supplying is 15 (out of 32), with 22 exporters that go only to 1 country.

This means that consumer in each market overwhelmingly face a different mix of supply, which is confirmed by the fact that the number of suppliers to a market ranges from 32 to 1. In no market all exporters are present and in all markets, except one, less the half of total exporters are present.

In eight markets, there are more than 10 suppliers (which should assure a significant competitive wealth of choice), but in 13 markets the suppliers are 3 or less (a sort of oligopoly).

The 56 competitors could theoretically meet couplewise in (56*54) / 2 = 1540 ways. Out of this theoretical maximum, there are 832 couples of competitors that actually compete for the same market (or more than one market). More than half of theoretical possibilities are actually present. In a totally globalised world, all exporters would meet in all markets. By contrast, the maximum width of exposure to competition is 10: Italy and Spain compete in 10 common markets and they are the couple which competes the most. Since Italy exports to 15 markets, it has to do with Spain two thirds of the times. Conversely, since Spain is active in 12 markets, for Spain, Italy is a competitor in 83.3% of the times.

Exporters present in many markets tendentially face a more diversified competition than exporters present in fewer markets. Italy is member of 43 competitors' couples. But other countries, present in markets where there the number of competitors is large, can take part to even more couples: Japan, for instance, is present in 53 and South Korea in 51. Spain is at 44.

More in general, exposure to pairwise competition ranges from 10 common markets to 1 (with 535 couples meeting only once, 297 twice, and 100 couples meet at least in 4 markets ).

Now let's turn to the issue of the relative strength of each competitor in a couple. If we take Japan and Italy, the former is "strong" without adjectives in 2 of the 8 markets where they compete: in those two markets Japan sells at higher prices higher quantities, making a lot more in terms of turnover. In other 5 markets, Italy is a "strong price undercutter", i.e. by making lower prices it sells more than Japan and total values are higher than Japanese ones. In the last market where they compete against each other, Italy occupies a "weak premium niche": it sells at a higher price than Japan, has less sales and lower total turnover. In other words, Japan products are more expensive than Italy's in 7 out of 8 markets and Japan is stronger than Italy in 7 markets (be it strong or strong premium niche).

More in general, out of the 297 couples that compete in at least two markets, the quality of their competitive relationship remains the same across all markets for 93 couples (31%), whereas it oscillates, according to the market, between two qualities in 132 couples, 53 among three qualities, 15 among four, only 4 among five, and no instances are for six.

You can see that this kind of analysis allows to highlight how a general competence in production and distribution is converted into relative competitive strength, with some variations due to the specificities of the target markets.

Counting across all couples, the most frequent relationship between two competitors in a market (61% of all bilateral relationships) is that one of them is a strong price undercutter and the other is weak premium niche, followed by the situation in which one is "strong" and the other is "weak" (30%), with the remaing 9% is represented by couples where one strong premium niche and the other is a weak price undercutter.

Cutting price is more frequently a winning strategy than betting on higher price (61% to 30%) with a small probability that both strategy are partially successful (9%).

Finally, we build a quantitative index of strength, through a two-step procedure. In the first step, we judge each couple of country, to highlight which is stronger, by counting how many times is "strong", "strong price undercutter" and "strong premium niche". Taking the percentage of such computation over the total number of markets where they compete and considering a threshold of 60%, we declare stronger if it is in those three categores for 60% or more of the markets.

Applying this first step to our example, of all 832 couples, 789 couples are unbalanced, with one country clearly stronger than the other (i.e. it has more markets in which is stronger). For instance Italy is stronger than Greece in all 7 markets where they compete.

Conversely, in 43 pairs of countries, the situation is more or less balanced (with prevalence of one over the other in less than 60% of the markets where they compete). For instance Italy and Germany share 8 target markets, in 4 italy is stronger, in 4 Germany is stronger.

In the second step, we count in how many couples a country is stronger than the competitor. According to our "quantitative index of competitive strength", Japan is said to be the strongest competitor in this SITC, since is has the maximum number of favourable unbalanced relationships (43 out of 789). Italy is next with 38; Australia, France and Spain share the same value (35).

6 countries never have even one unbalanced relationship turning to their favour and can be considered the weakest competitors in this product globally.

Since in such index we summed up three types of being strong, it's worth re-introducing the difference and highlight a "quantitative index of core competitive strength" where we rank competitors by the number of unbalanced relations including cases where prices, quantities and values are all higher than competitors' ones.

To cope with the multiple competitors that can operate in the same market, the most streightforward approach is to compute all couplewise relationships and build a second-level taxonomy, in which their comulative results are considered.

If a seller is weak in all relations, it's "weak" for the whole supply structure. If it is "strong" in all relations, it is strong for the whole supply structure. As you can see, you can obtain a fairly high number of intermediate cases, depending on the data and the actual number of competitors.

By the way, the simplifying assumptions that all sales, quantities and prices are different over the same market is corroborated in 224 out of 226 sale values (99.11%), and in 226 out of 226 (100%) of quantities and prices. Empirically, in this product, never are prices or quantities sold by two competing seller in the same market.

These results hold for the single SITC market we considered; you can experiment with the others, to verify the robustness of the abovementioned statements and a few key statistics and questions:

1. which is the most competitive country in the world, by counting in how many products is the country at the top of our "quantitative index of competitive strength"?

2. which pair of countries is engaged in the most competition against each other, based on the number of markets and SITC in which are both present at the same time?

3. in which SITC a country has a competitive advantage, by comparing its ranking in the "quantitative index of competitive strength"?

4. is price undercutting always the widest used strategy to achieve higher value of sales?

5. is there a relationship betwee total size of the market and the number of competitors (the larger the size, the more the competitors)? and with price (the larger the number of competitors the lower the price)? are these relationships better explained by taking into account the relative competitve strength of couples of competitors?

6. is there a relationship between competition relations as analysed here and the hierarchies in trade we investigated here in static terms, here in dynamic terms and for which we proposed a few trade policies?

In another vein, you might want to investigate in formal models how it is possible and sustainable that someone selling at higher price can sell even higher quantities than its competitors. You probably think that the issue of product differentiation and quality are crucial. Indeed you can explore how our discussion of vertical, horizontal and mixed product differentiation can justify those results. But further explanations are possible, including competitive strength arising not from the product itself but from earlier investment in supply chains, historical linkages to markets, power relationships, and special treatment achieved.

Dynamic and qualitative approaches to relationships among competitors

With all its richness, the previous analysis was making use of just three data (sales, prices, and quantities) in a specific year to judge the quality of the relationships between couples of competitors. But a strong competitor could well be identified looking at the time series of business results in one or more markets, vis-à-vis its competitors. A sharply rising market shares in all markets in which takes part would be a relevant signal in this direction. The dynamics of market shares could provide quantitative elements for qualifying competition.

You might explore a dynamisation of the six abovementioned roles, by taking the differences over time of prices, quantities and sale values.

But you can also remove the request of data availability to consider more qualitative and structural reasons for strength. In particular, by playing and interpreting our model of "Bidirectional vertical product differentiation with bounded rational consumers, innovation, advertising and finance" you can explore how investment in Research and Development for both product and process innnovation (including organizational innovation) can result in excellent product performance, improved value-for-money and in superior match to customer needs and tastes and how advertising (together with word-of-mouth through personal and social networks) can magnify those differences.

A strong competitor can be seen a well-funded firm that can afford to invest in a new market by aggressively price an extensive "product line" (with highly competitive "first price", standard and premium products), coping with financial losses in first years and building on success in supply chain networks (e.g. distribution channels, presence in shops in the most dynamic cities, key complementary service providers, etc.).

Total profits and profitability (as percentage of sales or invested capital) are key short-run and long-rung indicators attracting new funds from shareholders and external investors.

Being able to take large loans, based on creditworthiness and assets, to pursue an aggressive investment strategy is part of the competitive strength. But also small, more agile and innovative firms can hurt the "big fishes".

You can experiment issues of profitability and innovation by using our model of competing firms having widely different fixed and variable costs.

The choice of target market, being in terms of segment and in terms of country, as in our model called "You are an exporter", is extremely important.

More specifically, monitoring of the competitors and the capability of respond to their moves in a timely and winning way (either by imitation or constructive criticism) is a very important specific competence.

The quality of management, the coherence characterising the strategy, the tactics and their implementation are all key factors, the positive relationship with the workforce, the customer base and the regulator.

Sometimes, one element is enough to drive success and hurt competitors; in other cases, it's the inimitable mix of tacit and explicit knowledge-base upon which the firm draws its innovative, productive, marketing and supply chain competences, embedded in routines, that makes the difference.

Its valuable, rare, inimitable, and unsubstitutable resources lead to sustainable competitive advantage, as Jay Barney would put it.

Localisation and local networks and specificities can be the key - especially for small and medium enterprises in neo-Marshallian industrial districts, innovative milieux, and eco-cities / eco-neighbourhoods where to test, to improve and to export products whose features, including their environmental and social sustainability, match the megatrends of our civilization.

5. A few consulting venues from these analyses

If you are directly involved in managing a firm or to support policies at sub-national, national or international level, it might happen that you want to turn these analyses into actions. They can be helpful in the decision where to export, which price and quality strategy to adopt, or to support an export strategy for your institutional body (e.g. a Ministry for Industry, Economic Development or Trade).

We can join forces in:

1. finding and using much more updated and detailed data than those distributed now for free;

2. developing an assesment of current competitive strength and weaknesses, both quantitatively and qualitatively;

3. explore pros and cons in different positioning strategies.

For instance, we might explore the potential of being the first to export in a market, maybe in geographical proximity or not, so to build a temporary monopoly, reach a critical mass in sales and leverage these early advantages in a sustainable competitive strategy, for when other countries or firm will enter the same market.

A second exploration is to start from prices that would be profitable to verify in which markets they can be practised without being undercut by competitors of similar or better quality. Conversely, we can identify which markets would appreciate the superior quality of our goods; in other words, where we can be strong in the three abovementioned ways.

All this could lead to a ranking of potential export destinations, so as to select a few to qualitatively deepen.

More in general, we can assist managers and policymakers in their competitive strategies in the market where they operate or would like to operate. Just contact us!



[1] The data are a sub-set of the well-documented work by Feenstra et al. freely distributed here. We simply eliminated the heterogeneous flows to non-countries (e.g. "Asia NES") and those denominated in units different from tons, so to have a full internal comparability. This eliminated more than 200 000 rows where products sold were counted in units (e.g. "sold 24000 items"). Accordingly, total exports of a country is normally underestimated in our sub-set. But for the goals of our analysis, where competitive relations are at the centrestage, these cuts are not particularly relevant, the more so since we are just exhibiting how the special analytical methodology to elicit competitive strength can be apply to any data of suitable format.

Main page