經濟文獻對勾結形成條件的探討，認為參與廠商人數是一重要變數，理由為廠商人數少時，可減少組成勾結的交易成本。是故，理論上可預期的勾結組織最後的結果是由少數廠商所組成。相對的，公會由市場生產相同或相似產品廠商組成，廠商加入公會在我國又是具有法律的強制性，因此公會平台提供了(1)產品為同質性，(2)全部廠商參與，有利於勾結形成的客觀條件。與一般勾結，所不同者，公會主導 的勾結，預期其組成廠商人數將比非公會主導勾結的廠商人數要多。 本文利用公平會執法十年後所累積的95件個案聯合行為個案進行統計分析，獲得的結果是(1)十年來公會介入的聯合行為個案的確佔有相當的比例。(2)公平法施行前公會介入聯合行為的比例，與執法後公會介入聯合行為的情形比較，似乎並沒有減少的趨勢。(3)非公會介入聯合行為的個案，參與廠商的確不高，平均廠商家數為8.13家，相對公會介入個案平均廠商家數為345家，若剔除極端值，兩者數據分別為7.06與153家，仍呈現前者遠小於後者之結果。(4)非公會介入個案的廠商眾數為2家或3家，與文獻上研究的結果對照比較相當接近，證實勾結組織的參與廠商人數一般不會太多。

From the extant economic literature that has delved into the factors that contribute to what constitutes collusive behavior, the general consensus is that one of the major contributing factors is that there are only a small number of participating firms. This may be because the transaction costs of colluding is considerably reduced when there are not many firms. Therefore, the theoretically predicted cartel is typically made up of merely a few firms. Another contributing factor is that the participating firms are connected to a relevant trade association. Because all firms that produce similar products are legally obliged to join a relevant trade association, membership in a trade association is a condition that is conducive to collusion. Thus, the number of firms involved in collusive behavior that have ties with a trade association should be greater than the number without such ties. This study conducts a statistical analysis of the ninety-five cases of collusive behavior that have come to the attention of the Fair Trade Commission during the ten-year period of law enforcement and concludes that: (1) During the ten-year period, a considerably large percentage of collusive behavior has involved firms with links to a trade association; (2) When the percentage of cases of collusion where there are such links to a trade association is compared before and after the enactment of the Fair Trade Law, little decrease is found; (3) The number of firms that have participated in collusive conduct but are not connected to a trade association is not high, with an average number of 8.13; contrast this with 345, the average number of firms that have connections to a trade association. When the extreme values are eliminated, the figures are respectively 7.06 and 153, still indicate that the former are far outnumbered by the latter; and (4) The mode of firms partaking in collusion without connections to a trade association is 2 or 3, which is fairly consistent with the results from previous reported research. This clearly shows that, broadly speaking, there are likely not many firms that participate in a cartel.