Thursday 29 December 2016

Unrealised expectations of unintended consequences - Update

A few years back I started pulling together a list of reforms that were enacted despite dire warnings of unintended consequences. As Albert Hirschman (PBUH) pointed out, despite the focus of conservative/reactionary voices on unintended consequences, an equally important question was unrealised expectations - when a policy was enacted and what was intended just didn't happen. I would say shareholder empowerment as an area of public policy probably deserves some analysis on this point, but that's another argument.

I thought I'd update the list because the Bonus Cap is a great example that we really mustn't forget.  

Reform - the Bonus Cap - bonuses in qualifying financial institutions should be a maximum of 100% of salary, or 200% if shareholders approved it.
Predicted unintended consequences - base salaries would increase to compensate (mainly) bankers, banks fixed costs would increase significantly, financial instability would increase.
Actual consequences - base salaries only rose a bit for a few bankers, fixed costs therefore did not increase materially, and financial instability did not increase either.

Reform - July 2000 Amendment to the Pensions Act requiring disclosure of pension fund policy on social and environmental issues.
Predicted unintended consequences - would make trustees the target of single issue campaign groups, would lead to pension funds screening out various types of stocks with a negative impact on the UK economy.
Actual consequences - a few more SRI mandates, though no noticeable increase in screening, more asset managers add corp gov/SRI staff (a positive unintended consequence?), no evidence (none that I am aware of) of funds being targeted because of disclosure of policy

Reform - public disclosure of shareholder voting records (ok, not an out and out 'win' yet, but there's much more data available now)
Predicted unintended consequences - would make asset managers... er... the target of single issue campaign groups, would be a major cost for asset managers
Actual consequences - some media coverage of and trustee interest in actual voting decisions (ie more accountability), negligible impact on costs (based on feedback I have received), more analysis of investor behaviour (ie various surveys)

Reform - annual election of directors
Predicted unintended consequences - would increase short-termism, could lead to entire boards being voted out
Actual consequences - too early to tell? But no examples of whole boards being voted out, and is there any evidence of increased short-termism?  

A few end of year thoughts

A few random thoughts that kept coming up during 2016...

It is not the case that different stakeholder interests within the firm necessarily align, even over the long term. As we learnt recently, for instance, some stakeholders will actively lobby against others getting a formal role in corporate governance - we have seen shareholders (asset managers) tell the government not to let workers have board representation. That is not alignment of interests, it is one set of stakeholders asserting their primacy over others in governance.

Various different models of corporate governance put different weights on various stakeholder interests, and even within countries these weightings have changed over time. (The idea that shareholders were significant players, or "owned" companies, was not taken seriously in the UK for much of the last century). There is no "end of history" with one model dominant. The endless attempts to rework incentives and time horizons within the shareholder-centric model in the UK shows that it has flaws like any system and others might be as good if not better. Our corporate governance community has shown too little curiosity about other models and falls for its own propaganda about our system's strengths.

It is unlikely, to put it mildly, that positive performance on all ESG issues is correlated with positive financial performance. Therefore if management of ESG issues is *only* viewed through the prism of generating financial performance this is likely to leave some people very disappointed. Even in simple tactical terms, it is better to argue some ESG issues as desirable ends in their own right, rather than in terms of their instrumental value for the creation of returns for shareholders (often with wobbly evidence).

If all ESG issues are viewed from the perspective of shareholder value then this obviously undermines the legitimacy of issues which don't show a positive correlation. The ESG community even has its own language for this - "materiality". If an issue isn't financially material, even if it matters a great deal for other stakeholders, then it is delegitimised as an area of shareholder focus.

If the objective of all corporate governance activity is defined in advance as the creation of shareholder value (even over the long term) then this really shackles the ability of business to play a positive role for other interests, and narrows down arguments to a ridiculous degree. If, for example, the objective of executive pay "reform" is to incentivise the creation of shareholder value, then the creation of shareholder value in firms that adopt shareholder-focused pay schemes might be taken mean that executive pay is "working", even if intra-firm pay inequality is rising.

Overall, "win wins" may not be as common as we like to believe. There may be plenty of times when choices within corporate governance involve advancing the interests of one set of stakeholders whilst not advancing the interests of another, or even damaging them. A focus on shareholder value directs boards to decide issues where interests do not align in favour of shareholders. Since shareholders do not own companies, and rarely even contribute capital to them, it is surely worth exploring (again) why shareholder interests should predominate. If a pro-shareholder settlement within firms contribute to outcomes that society finds troubling (like rising inequality) we shouldn't be surprised if public policy seeks to override it.

Asserting the public's interests as shareholders is not a straightforward issue either. Individuals in modern capitalist economies have different interests - as worker (or employer), citizen, consumer and investor. There is no inherent reason why these should pull in the same direction. As a consumer I might appreciate cheap Uber fares, but as a worker I may fear how business models like theirs seek to undermine employment protections. As a shareholder I might value greater returns in part created through lower labour "costs", as a worker (and a pensioner) I prefer higher wages. It is broadly true that the public are (indirectly) the largest class of shareholders now, but this varies greatly by country (both the extent of funded retirement system, and the extent of domestic ownership of domestic equity). In addition, the extent of an individual's shareholder interest will vary by economic background. Richer people generally have more extensive shareholder interests, those on lower incomes may rely more on state pension provision than private. So focusing on the shareholder identity has a class aspect to it (a point I'll come back to).

The way that shareholder primacy, embedded in corporate law and governance codes, is enacted in practice arguably gives shareholders more than one bite of the cherry. Shareholders have limited liability, so have nothing at stake other than the value of their investment. They are also provided with control rights to ensure that they can keep management focused on their interests. And in practice management prioritises keeping shareholders happy - ensuring that earnings targets are hit even if it means delaying projects, prioritising buybacks over investment etc. The effect of focusing management attention on shareholder interests above all others means that in practice they even come at the front of the queue in terms of claims on the company - this is not how the public company with limited liability is supposed to work.

Those who assert that "political" intervention is usually counterproductive and we should leave it all to the market often overlook how much of what we take for granted was the result of previous political intervention. For instance, shareholders are generally too weak and insufficiently motivated to gain extra powers for themselves. Shareholder voting rights and corporate disclosure of information across the board has been mandated by the state because market pressure was insufficient to deliver it. More recently the state has stepped in to promote "stewardship" - an activity is supposedly in shareholders' self-interest. Markets, and market participants, need politics and politicians for the world they operate in to function.

Rather than denigrating politics and politicians, corporate governance could learn from it. Politicians in general seem to have a much better understanding that most decisions don't involve identifying obvious win wins, and settling on one ageless objective. Decisions involving lots of people and various interests are often messy - and temporary -compromises where at least one stakeholder doesn't benefit, or even loses out. This is true in Big P politics, we should accept the reality in the micro politics of the firm too, and see what we can learn from elsewhere.

We should much more careful with language about "unintended consequences", and the implication that fiddling about policy always mucks up. Aside from the fact there are positive unintended consequences, and that predicted unintended consequences often don't occur (I need to update this list - auto-enrolment & levelling down, bonus cap and salary/fixed cost rises), often claims of "unintended consequences" reflect woolly thinking about who is doing what. If I undertake action X to get you to undertake action Y, and you know I want you to undertake Y, but you choose instead to undertake action Z, is your action Z an unintended consequence of my action X, or an intended (by you) consequence? Show your working...

Trying to redesign performance-related pay is a fool's errand. We keep failing to get the targets right. There is evidence that extrinsic reward drives out other motivations. Making people wait for reward - the aim of "long-termism" advocates - reduces its value in the eyes of most recipients (so rem comms make it larger to compensate). Variable pay has also been the source of most growth in executive reward. Our system is a mess that few really believe "works" and the emphasis on performance pay is at the root of may of our problems. I think this is a very difficult problem, and it's very unlikely, in my view, that it will be solved by more of the same.

CG/RI is shot through with the biases of those that work within the field. Hence the priorities reflect the sort of wealthy liberalism that has also dominated politics. Perhaps it's just a coincidence that asset managers focus more on paying execs more like finance industry players, and making boards more meritocratic, and very little on fair pay and voice for workers. Or maybe it reflects what economic background they come from, are part of, and expect to remain in.

Saturday 3 December 2016

ASOS AGM action!

This week saw some great campaigning by the GMB as part of their ongoing battle against poor working practices within ASOS, which has had a particular focus on the distribution centre in Barnsley run by XPO. The GMB had a "Catwalk of Shame" outside, and a workers' annual report was handed out to shareholders. On the same day there were also claims that ASOS may not be playing by the rules in terms of pay for new warehouse staff.

Interestingly, on my turf, the company was also hit by a sizeable vote against executive pay, with 33% opposed. This generated further negative media coverage. The vote against was also double the level of opposition (16%) at the 2015 AGM, so there might be something significant going on. Notably, ASOS has not commented on the size of the vote or how it will respond. As a reminder, the UK Corporate Governance Code says (E.2.2):
When, in the opinion of the board, a significant proportion of votes have been cast against a resolution at any general meeting, the company should explain when announcing the results of voting what actions it intends to take to understand the reasons behind the vote result.
In practice, larger listed companies seem to make statements when they get votes against of about 20%+. ASOS is AIM-listed so it does not need to follow the Code. Nonetheless you might think that given all the adverse commentary - and the scale of the vote - it would be a sensible thing to do. And actually ASOS does make commentary about shareholder support for remuneration in this year's annual report (page 47):

At the AGM last year, 84% of shareholders voted in favour of the Directors’ Remuneration Report, providing an important level of public accountability for the Board with the suitability of our remuneration policy and its implementation. We hope that you find this year’s Remuneration Report equally informative around how ASOS leadership is remunerated, and some of the changes that we have made during the year. I look forward to seeing shareholders at the AGM, and hope that I can count on your continued support on our pay arrangements.

So it looks like they may have something to say in future reporting.

Finally, we can dig into the voting data a bit. One thing that looks likely is that the company's major shareholder, Danish retailer Bestseller, did not oppose the remuneration report. According to the list of major shareholders disclosed on the ASOS IR site, Bestseller owns 23m shares. And according to the RNS statement on the AGM, just under 14m were voted against the remuneration report. I think it's likely Bestseller either voted all of its shares or none.

If Bestseller DID vote for the remuneration report, then a very large majority of the remaining shareholders who voted will have voted against. The RNS statement shows a bit under 28m shares supporting the remuneration report, if you take 24m out, you are left with 4m in favour and 14m against.

Just to complicate things, it's worth registering that former ASOS CEO Nick Robinson owns 7m shares. Which means that it does not look possible that both he and Bestseller voted their whole holdings (23m + 7m) in favour of the remuneration report. I suppose it's possible one or both has sold down a bit, but it looks a bit strange on first glance. If Bestseller didn't vote at all, but Robinson did, and voted for, then the level of independent shareholder opposition must have been around 40%.

It's too early to get any public voting data on this AGM, but looking back to the 2015 AGM, we can see that Baillie Gifford, the largest shareholder after Bestseller, voted against the remuneration report. (PDF here, if that doesn't work then look for Q4 2015 voting disclosure report here - you will need to accept terms and conditions). You can also see that ASOS was part of Baillie Gifford's engagement in the same quarter (look at the Engagement Report for Q4 2015). My gut feeling is that, given the increase in opposition this year, Baillie Gifford probably took the same position, but that is just a guess.

We can also see that one of the UK's local authority pension funds - West Yorkshire - also opposed the remuneration report at the 2015 AGM. See page 119 here, from the fund's voting disclosure site. Here's the blurb explaining the vote:

For Nick Beighton’s promotion to the role of Chief Executive, the Committee determined that the annual base salary level should be set at £550,000, which is £50,000 higher than his predecessor, and the annual bonus opportunity increased from 100% to 150% of base salary. The increases have not been adequately justified. Likewise, to secure the recruitment of Helen Ashton the Recruitment Committee bought out a proportion of her current long-term incentives from her existing employment, by making a one-off cash payment of £204,000 and a grant of a long-term incentive award under the ASOS Long-Term Incentive Scheme, worth £340,000 as at the date of grant. The awarding of a cash payment on recruitment, without any performance conditions attached, is not considered appropriate. 
I will keep digging around for info on how shareholders have voted this year and last, and why. I will post up what I find.

So, to sum up, we have a high-profile retailer whose working practices in one of its warehouses have been subject to critical media coverage. We also see, despite a large chunk of shares accounted for by insider ownership, the same company experience significant shareholder opposition over corporate governance concerns at multiple AGMs. Remind you of anyone?