Are large firms really successfully lobbying for low tax rates?

My last attempt at a review of recent academic literature on a corporate tax income topic was well-received, so here’s round two in the “Are firms really…?” series.

This time, I’ll take a look at another paper out of Oxford University in the context of an existing body of literature. Here, we are looking at the topic of whether large businesses are successfully lobbying politicians for low corporate tax rates.

Now, the political economy literature is torn on the matter. On the one hand, the “political power” literature holds that, indeed, large firms become politically influential due to advantageous positioning, political access, abundance of financial resources, etc. On the other hand, the “political cost” literature holds that firm size comes with increased public attention and thus relatively higher cost levels incl. taxation.

The former hypothesis is probably well-known to most readers, given the popularity of “business influence” or “regulatory capture” stories in academia and the media. A recent example is a story from Vice, titled “Bigger corporations are making you poorer” (, citing an interesting study by economist Simcha Barkai at the University of Chicago, which provides a link between industry concentration and firm profits, indicating a size effect on the ability of companies to extract value from the economy. And that’s just the latest in a long line of research on the firm size-influence connection. This the mechanism at play in the “political power” hypothesis.

The latter, however, is probably not as widely known, even though a noteworthy body of literature indicates that this is actually the theory more aligned with reality.

A study from February 2016 ( by a group of researchers at the University of Mannheim reviewed 46 papers dealing with the political cost/political power forces. Performing a meta-regression on the existing evidence, they conclude:

This analysis is interesting because the empirical results across primary studies are by no means clear-cut and no tendency can be observed towards either the political cost or political power. Generally, our findings confirm the political cost theory. (my emphasis)

In short, “it’s uncertain but we’re leaning towards the political cost hypothesis”.

Their overall findings conclude that, on average, a ten percent increase in firm size, leads to an increase in effective tax rate of between 0.5 and 1 percent.

However, just as I remarked in my previous post, the wide variety in methods, data and analyses across the literature make it difficult to conclude definitively on the superiority of the political cost theory. The authors specifically address this point, noting that key definitional variations of firm size, of effective tax rates and of the sample period significantly affect the estimation results. For instance, the size-cost connection is significantly larger when measuring size by total sales, compared to total assets.

There’s also the question of concluding “on the average”, which is what meta-regressions like this can tell us about, as opposed to “on the margins”. It may for instance be that an “on average” increase in ETR by firm size hides significant variations by firm size level, e.g. medium-sized firms paying relatively more than small-sized firms but mega firms paying relatively less than large firms.

We must also pay attention to the geographical coverage of the survey. Of the 46 papers in the Mannheim study, the vast majority sample only US firms, with only a handful focusing on Europe. Thus, there are questions of generalisability. (I note here that the studies that do sample European firms generally seem to have noticably lower correlations between size and ETR than US sample studies, perhaps indicating that ‘largeness’ in Europe is less likely to produce political costs than elsewhere – but definitive evidence remains elusive.)

That leads me to the recent Oxford paper (, released in April 2016, which provides new and relevant evidence on the topic.

How so?

The Oxford study contributes to the geographical literature gap by utilising rich administrative micro-data for firm size and tax rates across German regions. The intra-national comparison provides an interesting and novel ground for study of these dynamics.

Moreover, the Oxford study looks at the relationships between firm concentration, a relative measure of largeness, as opposed to studies of absolute firm size. The underlying notion is that relative size advantage may be more important than absolute size in the political game, and/or that relative firm concentration within a community may foster collective benefits for businesses.

Finally, the paper studies the link with German local business tax rate choices, i.e. regional statutory rates, as opposed to effective rates, which most of the literature utilises. ETRs are notoriously subjective, difficult to measure, and a source of uncertainty in the literature (as noted above), so there is sense in using statutory rates. Of course, ETRs are an individual firm-level measure, whereas headline rates apply (at least in the first instance) to the community of local businesses as a whole. Each may be useful to study, but in the context of this paper, focus on statutory rates match the focus on firm concentration (also a community-level measure).

Based on the data, the Oxford crew carry out a number of econometric statistical regressions, arriving at the conclusion that:


(…) we find that higher levels of firm concentration in German communities are associated with lower local business tax choices. The effect turns out statistically significant and quantitatively relevant and prevails in various sensitivity checks. (…)

The results confirm prior descriptive and anecdotal evidence, e.g. suggesting that the German cities of Wolfsburg, Ingolstadt and Ludwigshafen, which host the head-quarters of leading car and chemical manufacturers Volkswagen, Audi and BASF, set considerably lower local business taxes than otherwise comparable jurisdictions with a more even firm size distribution.

The German evidence for the presence of a political power dynamic is a notable addition to the existing literature, which seems to have favoured the political cost dimension. Specifically, the Oxford authors find that a 0.1 point increase in the local Herfindahl index, a conventional measure of economic concentration (measured as the sum of squares of the market shares of firms in the defined community), reduces the local business tax rate by 1.7% on average. Again we must note all the caveats of above: definitions, “on average” vs “on the margins”, etc.

Although this seems contrasts with the Mannheim review, the Oxford study may actually align relatively well with this existing literature given its geographical emphasis (Germany). As noted above, European-focused studies seem to provide weaker evidence for the political cost hypothesis than US-sample studies. In any case, it is further evidence relevant to the sum of knowledge on the topic.

I don’t purposefully set out to end all of these blogs with the same conclusion: that there is competing evidence, all of which is valuable on its own, but provides no firm, universal takeaway. But that is where we seem to have arrived. The political cost theory does seem to have slightly favourable support, but only in the specific context of existing studies (e.g. US-centric.)

However, I do want to stress that we should take the particular context of each study into account when evaluating a coherent body of knowledge. Often, I think we do academic research a disservice by seeking easy generalisable answers from it, or falsely ascribing universal laws to it, when in reality it rarely, if ever, meets such lofty goals.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: