In All Things, Moderation – Including Outside Firms: For an in-house lawyer to get to know and work well with a handful of firms seems entirely reasonable based on the data

Saturday, November 28, 2015 - 14:38

For decades, general counsel of U.S. companies have strived for the right number of external law firms to retain. Very broadly, three camps have emerged: the Convergers argue for a small number of law firms to handle most of the matters and receive most of the company’s spend; the Coursers believe in “horses for courses” and for the most part let their in-house lawyers retain the number of firms they think best for their matters, even if that means hundreds of law firms on the department’s retention roll; and in the middle, the Compromisers advocate using some primary firms for much of the work and then a moderate number of secondary and tertiary firms, as needed, for specialized services or other reasons.

Members of each camp claim superiority for their approach, but there has been a paucity of empirical research regarding who has the better argument in terms of managing external legal spend. Ideally, we would like to see data for a group of law departments on their external spending stated as a percentage of a company’s revenue and match that ratio against their concentration of spending by law firm. But such a data set has not been compiled, despite years and years of benchmarking surveys. General counsel simply don’t want to provide sensitive data, such as how much their law department spends on each law firm retained.

If we had detailed information from departments on spending per firm, it would be ideal to calculate Gini coefficients. A Gini coefficient would tell how concentrated the spending is for each law department. After all, a department might have 15 firms paid more than $10,000, but two of them were paid 80 percent of the total fees of the entire group. The department would have a very high Gini coefficient because its distribution of spending was highly unequal among the firms paid (Gini coefficients are commonly used to show how income skews toward the very few high earners – the so-called “1 percent”). So the debate continues with each camp trumpeting its anecdotes and no one able to analyze the metrics of a good data set.

We can, however, get part of the way there. Short of the ultimate data set of large numbers of U.S. law departments and their external spend breakdown (plus revenue), there is some data that can get us part way. This year, General Counsel Metrics, which conducts the world’s largest benchmark survey of law department staffing and spending (sponsored by the executive recruitment firm Major, Lindsey & Africa) added a new question. It asked respondents to estimate the number of law firms their department had paid more than $10,000 dollars during FY 2014. The threshold served to drop out firms being paid for leftover FY 2013 work and such firms as local counsel who hardly register on the law department’s budget for outside counsel.

Through the summer, 341 respondents had provided their estimate of the number of “10K” firms they had used, ranging from a single law firm to 516 firms. Twenty-five percent of the respondents reported five or fewer 10K firms; 25 percent of them reported 25 or more 10K firms. The average over all 341 companies was 26, while the median – the middle figure when the numbers are sorted from highest to lowest – was 10 firms.

Even such purely descriptive data yields some insights. For a large group of mostly U.S. law departments – a group with median revenue of $900 million and average revenue of $3.5 billion – two dozen firms or so handled nearly all their legal work. Companies with revenue of $1 to $2 billion, roughly in between the average and the median revenue of the GC Metrics U.S. set, would typically have 5 to10 lawyers. If so, that would amount to something like two to three law firms per in-house lawyer, which would probably surprise many people as quite low. For an in-house lawyer to get to know and work well with a handful of firms seems entirely reasonable.

But we can go farther with this data and graph it. First, let’s narrow our data set to U.S. companies that reported at least a $500 million in revenue, since below that mark, benchmark metrics can become a bit anomalous. Of the total group of 341 companies, 281 indicated that they are headquartered in the United States. (The summary data for them is almost the same, interestingly, other than the median rising slightly from 10 to 12 firms.)

Now, let’s calculate for each company (164 U.S. companies of more than $500 million in revenue) how many 10K firms they used for each billion dollars of their revenue. For example, an $800 million company that used 14 firms would be at 17.5 10K firms per billion (14 10K firms divided by 0.8 billion in revenue). From that we can create a plot, included here, that shows 164 points, one per company, at the spot above the company and parallel to the left axis figure for the company’s number of 10K firms per billion of revenue. 

But the plot actually adds more insight. First, it has sorted the companies from the one with the lowest revenue (close to $500 million) on the left to that with the highest revenue (more than $60 billion) on the right. This lets us start to see camp membership, that is, whether law departments use more, fewer or approximately the same number of law firms as their revenue increases.

Second, to make that last empirical finding clearer, let’s add a linear regression line. Such a line minimizes the distance of each point from the line; assuming a linear relationship, it clearly shows that the larger companies grow, the fewer firms they hire per unit of revenue. The decline is roughly from 15 firms per billion to 10 firms per billion, a drop of one-third. As for camp membership, none of the three can claim a preponderance of adherents.

Let’s dig deeper. Are there differences between industries in the number of law firms they use or how that number changes with revenue? Yes. The plot with four lines shows U.S. companies with more than $500 million in revenue doing business in four industries: health systems (11 companies), manufacturing (38), retail (11) and technology (21). Each of them shows declining ratios with increasing size, but this limited amount of data points doesn’t support a more precise conclusion than that. You can see that the top line on the right, technology, stays fairly flat whereas the lowest line on the right (dotted and green for retail), drops the most.

We can summarize our findings, based on this reasonably extensive data set, as follows: The Convergers are in a distinct minority, but so are the promiscuous retainers, the Coursers. Instead, the most prevalent pattern appears to belong to the Compromisers, where a modest number of law firms dominate the spend of U.S. law departments and a very modest number of firms per in-house counsel who manage external firms.

As in most things, it seems, moderation once again holds true.


Rees MorrisonA principal at Altman Weil who also leads General Counsel Metrics, which offers the largest benchmark report of law departments ever done.