Daniel R Bereskin QC, Bereskin & Parr LLP
Daniel Bereskin QC at Bereskin & Parr explores the problematic nature of “likelihood of confusion” when it comes to assessing unfair competition in the trademarks space.
Unfair competition is a continuum. At one extreme, there is outright fraud or counterfeiting, where a trademark owner’s trademark is applied to a good with the immoral intention of deceiving buyers into the belief they are purchasing the trademark owner’s good. At the other extreme, there is conduct that is both harmful to the economic interest of the trademark owner and is outside the reasonable bounds of honest competition. Linking these extremes is the fictitious legal construct called “likelihood of confusion”.
The argument of this article is that instead of basing injunctive relief on proof of “likelihood of confusion”, the court should assess whether or not, in any given case, the defendant’s conduct is sufficiently unfair as to justify legal sanction, without guessing what the relevant public thinks at the point of purchase. The court should rely on well-established factors for determining if the impugned conduct amounts to unfair competition of sufficient magnitude to justify an injunction, taking into account the public interest in freedom of competition and freedom of commercial speech. This does not exclude evidence of actual confusion or likelihood of confusion, but makes confusion a factor, and not necessarily the decisive factor, in assessing whether the extraordinary relief of an injunction is appropriate.
For example, there are cases in which the respective trademarks or trade names are both similar and descriptive, leading to a certain degree of confusion. Accepting some confusion may be preferable to granting an undue scope of protection to a descriptive trademark or trade name: see Office Cleaning Services Ltd v Westminister Window and General Cleaning Ltd (1946) 63 RPC 30. Another example is a functional trademark. Even in cases where actual or likely confusion has been proved, if the trademark is found to be primarily functional, courts do not enjoin the use of the impugned trademark: see Kirkbi AG v Ritvik Holdings Inc. 2005 SCC 65 (imitation of Lego brick).
In short, it is proposed that the likelihood of confusion test ought to be supplanted with a system that carefully balances the rights of traders and the rights of the public to freedom of competition and freedom of expression. Worthy objectives are to reduce the cost and complexity of trials and relieving courts of the difficult task of probing consumer minds. A number of scholars have addressed this important topic.
In an article entitled Confusion Isn’t Everything ((2013) 89 Notre Dame L. Rev. 253) William McGeveran and Mark McKenna discussed the manner in which confusion is adjudicated, and in particular how the concept of likelihood of confusion has expanded in the US over the decades since the time when trademark injunctive relief was more clearly predicated on proof of deception (which in general required proof of diversion of trade). Their article concludes as follows:
Confusion isn’t everything in trademark law. It is merely a heuristic to guide courts in identifying situations where ambiguity about the provenance of goods or services interferes with the market and distorts the overall information available to consumers. It is a means to an end. In many ordinary trademark cases, focusing on consumer confusion continues to work well. But at times, that focus undermines the very competition and communication values that justify trademark rights. We need straightforward limiting doctrines that put confusion detection back in its place – not as the purpose of trademark law, but as its tool.
In Taking The Confusion Out Of “Likelihood Of Confusion”: Toward A More Sensible Approach To Trademark Infringement ((2012) 106:3 Northwestern University L. Rev. 1307), Robert G Bone makes the following observation concerning the likelihood of confusion test:
The test is a mess. It produces bad results, is doctrinally incoherent, and lacks a sensible normative foundation. It chills socially valuable uses and facilitates excessively broad expansions of trademark law, and it includes factors that make no sense as predictors of likely confusion. Most importantly, the test erroneously assumes that consumer confusion is enough to warrant liability without regard to the harmful consequences or the moral wrongfulness of the defendant’s conduct … Trademark law has suffered under the likelihood of confusion test for far too long. The problems run deep and ad hoc tinkering with factors will not solve them. The test needs a complete overhaul. A properly functioning trademark system depends on it.
Since the 1970s when survey evidence generally became admissible, lawyers have relied extensively on trademark surveys to prove or to disprove likelihood of confusion, but that habit is now under attack. In the UK, surveys currently are not admissible without the prior approval of a judge, whose approval is far from automatic. The leading trademark case is Interflora v Marks and Spencer (1 and 2), (1) (2012) EWCA Civ 1501 and (2) 2013 EWCA Civ 319.
In Interflora it was decided that:
a party may conduct a pilot survey without permission but at its own risk as to costs;
no further survey may be adduced without the court’s permission and disclosure of the pilot survey; and
no party may adduce evidence from survey respondents without having the court’s permission.
Also, even though a survey technically may be admissible, the court is to evaluate at an early stage whether it would be of “real value” and whether its likely benefit would justify its costs.
In the US, trademark surveys are thought by many litigators to be highly influential, but several studies have questioned this perception. In a paper entitled An Empirical Study of the Multifactor Tests for Trademark Infringement (95 Cal. L. Rev. 1581 (2006)), Barton Beebe reviewed 331 published cases and found that only about 20 per cent of the cases referred to survey evidence at all, and the results of the survey were consistent with the decision in only 7 per cent of the cases. He concluded that survey evidence in practice is of little importance.
In a paper entitled The Role of Consumer Surveys in Trademark Infringement: Empirical Evidence from the Federal Courts (14 J. Bus. L. 1013 (2012)), Bird and Steckel studied 533 cases and found that in only 16.6 per cent of the cases, survey evidence was referred to in the opinion. They concluded that consumer surveys are not particularly useful even in likelihood of confusion cases.
Prior to the 1970s courts decided many trademark cases without the assistance of a survey, and their decisions seem as reasonable by today’s standards as decisions in which trademark survey results have been admitted. Further, even if it is true that in some cases a proper survey is a better means of proving likelihood of confusion than the judge’s experience and common sense, it is often the case that if the results of the survey disagree with the judge’s own assessment, the judge will find ways to reduce the weight of the survey, if not dismiss it altogether. Ultimately, it is up to the judge to decide whether the probative value of the survey outweighs its cost.
So, the question is, how should a court determine whether or not the impugned conduct ought to be enjoined? In the US, each federal circuit has developed a set of factors that are to be considered in assessing this. The most famous case is Polaroid Corp. v Polarad Electronics Corp (287 F.2nd 492 (1961)) where the following variables were identified:
… the strength of his mark, the degree of similarity between the two marks, the proximity of the products, the likelihood that the prior owner will bridge the gap, actual confusion, and the reciprocal of defendant’s good faith in adopting its own mark, the quality of the defendant’s product, the sophistication of the buyers. Even this extensive catalogue does not exhaust the possibilities – the Court may have to take still other variables into account.
Note that these eight factors are not necessarily the only ones to be considered.
In the author’s view, determining whether or not a defendant should be enjoined requires a careful identification and weighting of all the surrounding circumstances including the effect on competition and free speech. De-emphasising proof of “likelihood of confusion” is against the instincts of many trademark litigators, but it is submitted that engaging in costly trademark surveys that may not survive careful judicial scrutiny is not the answer either.
In conclusion, relying unduly on surveys to prove or disprove likelihood of confusion in trademark litigation often is not a cost-effective strategy. It is better to put the emphasis on other factors pointing to unfair competition. That said, a survey of modest scope and cost (ie, one that cannot scientifically be extrapolated to a much larger population) can be of value to the court if the survey is designed and conducted fairly, measures the right issue, the importance of the survey is not overstated, and the survey is only one of a number of pieces of evidence relied upon to prove or disprove unfair competition.