Sean Park Portrait
Quote of The Day Title
In the beginner's mind there are many possibilities. In the expert's mind there are few.
- Shunryu Suzuki

On compensation and outcomes

Many headlines recently have focused on the continuing discussion surrounding reforming pay in the investment banking industry – including via legislative fiat. And particularly in the UK, this has been juxtaposed against the sordid disclosure of expense shenanigans by MPs (to the glee of many bankers it would seem.)

I hesitate to wade into this debate with my observations or suggestions: it is complex, has been studied at length by many academics and professionals much more qualified than I, and almost certainly does not lend itself to simple, mechanistic solutions. (Which is why I am highly skeptical when any politician suggests that more detailed laws or rules are the answer.) However I’m extremely frustrated with the lack of ‘outside the box’ thinking in this debate – not because it is necessarily where the answers will be found – but because unless one steps out of the frame of conventional wisdom you are almost certainly condemning yourself to come up with sub-optimal solutions. Especially since some of the fundamental tenants that are taken as gospel might be wrong. So with the disclaimer that I don’t pretend that the following suggestions are not without their own problems, I think they are worth considering, if for no other reason than to change the frame of the debate. Indeed I first thought of writing this post 3-4 years ago but it always seemed kind of pointless as at that time, there was absolutely zero appetite for considering any other compensation paradigm in the industry. I mean why would you, right? It was all good. Happy days.

I would suggest that most investment bankers (and MPs) are not paid enough. Enough base salary that is. But wait a second I hear you saying (well perhaps not today but certainly a couple years ago), the strength of the system is you can make a lot of money sure, but only if you perform. Who could argue with that? Well I will. For two main reasons. And I think if my suggested approach had been the norm in the industry, we would not have seen the same degree of egregious and venal behavior and may even have avoided some of the worst excesses.

  1. First is the problem of credibility and the fit-for-purposeness of the bonus. In a system when even average employees in average years get half or more of their total compensation in bonus payments, and top producers in top years get most of their total compensation in bonuses, the idea that bonuses are entirely discretionary and variable is delusional. Paying out zero bonuses – even in poor years – is a nuclear option – the credibility of the bonus as a completely variable element of compensation is significantly if not completely undermined. So in effect the variability (both to the upside and the downside) is in reality highly damped. ie It starts to look like mostly fixed costs; and yet crucially for the employee it is not and so you have the worst of both worlds: the employee does not benefit from the security of a larger fixed remuneration, but the company effectively is committed to paying a high quasi-fixed cost. It’s worth taking an example: imagine a banker with a base salary of £100,000 per year who over the past 5 years has received bonuses ranging from £90,000 in a really poor year for the firm and his activity to £250,000 in a great year. On average his bonus’ have been £160,000 over this time. I would suggest this is a dumb way to pay this professional. Rather his salary should be around £200,000 and his bonus should have been beetween 0 and £40,000 every year except for the stellar year when perhaps it should have been £100-200,000. ie The norm should be c. 0-20% of salary. By taking this approach – given that the employee will value the greater certainty of the second construct, the company will be able to pay this banker less on average and yet have a more satisfied and aligned employee. As long as the first question asked when considering the bonus of an employee continues to be “what did he/she get paid last year?”, the robustness of the bonus process will be somewhat of a farce.
  2. Which brings me to the second point. I can already hear some of you screaming out that this would just encourage a bunch of freeriders and goes against the core tenants of paying for performance. Well no and no. First of all if your employees are being paid these kinds of salaries and choose to freeride, they should be sacked and then you should be sacked for hiring the wrong people. Without exception, the very best professionals I have met, at all levels of the organization, do not work more or less hard because of the money: they work hard because they want to succeed, because they are passionate about their work. And they expect that if they are successful that the money will come. Any of you who have ever worked on a trading floor, do this thought experiment. To the best of your knowledge, what was the correlation between those that worked the hardest, put in the most effort, and those that were paid the most? Clearly, and especially in big companies, there will always be people ‘along for the ride’, trying to live off the efforts of others; often knowing they probably won’t get paid the most but they will certainly work the least. Worse, if push came to shove most people in most organizations can easily identify most people like this. By paying salaries that accurately reflect the value of the opportunity and the median pay for people with the skills to capture the value of the opportunity – I would guess that you would actually have fewer freeriders. The company couldn’t afford to tolerate them. They would have to remove them.

Adjusting the compensation paradigm towards a higher fixed component and a more sensitive and truly variable component (including averaging payouts over 4-7 years, including the possibility of down – ie malus – years) would also go a long way in improving risk management by going with the grain of human nature rather against it. Behavioral Finance 101 applied to human resource decisions. Again I suspect that had banks (a) had proper internal transfer pricing of cost of capital and liquidity and (b) a more salary-based compensation policy perhaps we could have avoided the most toxic behavior that developed in the structured credit market. A lot of the original ‘mark-to-market’ profits that were generated out of thin air (upon which very real bonuses were paid out) essentially were nothing more that arbitraging long term risk against overnight funding and the inability of the bank to charge appropriately for the capital used to allow these trades. Indeed the extreme complexity of many of these structures was imo just plain old misdirection – like a magician’s illusion. It made the profits seem plausible – after all it was really clever stuff – and created an illusion in management’s eyes as to the real source of the profits and the associated risks. Plus these very clever – and increasingly rich – people also, like most of us, liked to show off a bit and so – like musical virtuoso’s – competed to produce the most elegant and complicated embellishments, just to prove what could be done. If the average MD in this business had had a salary of say £500,000 with an expected bonus of £0-750,000; rather than a salary of £125,000 and an expected bonus of £500,000-2,000,000 you would have produced a very different, more healthy, set of behaviors I believe.

To conclude, I’d like to reiterate that I am not suggesting that this is a perfect solution – it clearly has its own problems – but rather I am suggesting that it is a much better system than the existing one, and would produce better and more balanced economic outcomes over time for all stakeholders (employees, shareholders, taxpayers.) So as a roadmap to the great and good who are currently tasked with revisiting pay in banking I would leave them with these three suggestions:

  • pay bankers higher basic salaries
  • create bonus/malus schemes that match the underlying business and asset structure (some banks have already started to do this, they should be applauded)
  • encourage smaller organizations (in particular for high risk/reward activities) where it is much easier to align and monitor the interests of all stakeholders

Late breaking news (FT, 22may09):

Morgan Stanley is changing its compensation scheme by de-emphasising the year-end bonus and increasing executive salaries as the Obama administration prepares to introduce a set of broad reforms aimed at changing Wall Street’s pay incentives.

The firm’s board of directors approved an increase in the base salaries of several of its top executives, while at the same time reducing the end-of-year bonus, according to a regulatory filing. Other banks, including UBS and Credit Suisse, have moved in a similar direction.

Wow. Watch them all fall in line now. It’s all about not being different than your competitors. Great to see this idea being put into practice (even if it did take a Level 5 financial storm to do it…) When I first suggested this idea to my peers and bosses 5+ years ago it’s not an exaggeration to say they looked at me like I had horns growing out of my forehead…

Update (27 may 09):

Citi, BoA may raise base pay for investment bankers (Reuters)

Well that was quick! Didn’t know Vikram and Ken were readers. ;)

Related articles by Zemanta
Reblog this post [with Zemanta]

Post to Twitter Tweet This Post

Permalink||Comments|Filed under 'Banking, Management'
  • martin
    The main problem with your point #2 about freeloaders is that it's actually very hard to fire people in many places (even in the US) for poor performance. It takes the organization months and months of arduous, self-defeating processes. Arduous because it's like: give a bad review, then verbal warning, three written warnings, negotiate "no you can't get more training because it's not going to make you smarter", etc. Self-defeating because once one gets a bad review or warning, there is *zero* incentive for them to improve.
  • You raise a great point Martin. Rather than try to reinvent the whole social and legal framework around workers rights, I would suggest it might be easier - politically and otherwise - and probably more socially fair and desirable to carve out exceptions for highly paid professionals.

    Indeed while I think labour laws in many countries, especially continental Europe are too inflexible and over-protective and result in a less good outcome for the economy, I do have much more sympathy when protecting a worker making £30,000/year, even someone making £80,000 a year. But for someone making £200,000 or more, I think it is reasonable that there is more employment risk. Which by the way they should be in a better position to self-insure against.

    Clearly, it wouldn't be as easy as that to implement such a carve out: at what level, how, for whom, under what circumstances, etc.? But nor should it be impossible. And it should even play well on the campaign trail.
  • One thing that gets overlooked by most banks in terms of compensation is the interrelatedness of each separate business unit. I see most large banks as a combination of investment banking, trading, and private client. Each one depends on each other unit to do its job. Brokers historically have been paid a variable salary but often times (although not as much lately) depend on the quality of the deals being underwritten by the investment bankers, and the quality of execution by the traders. The bankers depend on the brokers to bring in client capital to buy up the deals and the traders to move those deals to the markets. The traders depend on both to give them flow.

    This brings about enormous moral hazard abound within the organization because each unit is an arms length away from the customer itself. It creates a lot of tension within the organization and disdain from the customer itself. If the bankers are loading up the traders with too many overvalued assets, they are not adding value to the organization. If the brokers are not bringing in enough money then the bankers can't be competitive in the underwriting of deals, and so on. There must be an element of compensation that brings the organization together, because in turn this will align the organization's compensation with the needs of the diverse client base, not one that is simply preying off of other business units' future profits/losses.
  • Totally. Don't get me started on this.

    But basically, yet another (good) reason to increase salary component of total comp.
  • Agreed. Base Salary plus bonus based on total firm profits.
  • Chuck Farley

    Another aspect of this issue that should be addressed is the implicit optionality of finance compensation. Though not exactly "owners," all employees have a role in the direction and decision making of the firm. If these same employees are credited with cash in the good years (like a dividend), but not debited in the bad years (your "quasi-fixed cost"), the compensation starts to seem like equity! (ownership + limited downside + unlimited upside)

    Equity is holding a call option on the firm's assets, an option written by the debt issuers. As most of us know, (call) options are more valuable with increased volatility. As total compensation becomes more equity / option -like, employees have an incentive to swing the firm's value wildly via projects that are very risky relative to the firm's cost of capital. ("I don't have to pay if the project chokes, but I get perks if it succeeds.") This is not "good" entrepreneurial risk, but simply gaming capital structure for some serious moral hazard.

    So yeah, bankers should make larger base salaries and less bonus.
  • Sean,

    I'm going to disagree twice in one day. Your argument has a lot of common sense running through it, and it's right in so far as that's how things probably need to work inside of existing bank style corporate governance; but it doesn't address the underlying issue that what's really needed here is a system that allows, or rather encourages, staff to be entrepreneurial, which is hard to do inside the hierarchical employer-employee relationship. So... if we accept that banks will have a fundamentally broken relationship with their staff then you're right, but should they be fixing root cause or symptom?

    I was lucky enough to attend an inaugural lecture by Professor Andrew Pendleton whilst I was at a summer school at the University of York last year. He argued that the cost of capital of organisations can be modeled against the variable compensation of employees (more variable = lower cost of capital). Not very CAPM, but then we all know that CAPM is broken. I was intrigued that it provided a good explanation for why successful startups succeed (though clearly it conveniently ignores why startups with similar capital structure fail). Anyway, you might find the material at http://www.york.ac.uk/management/staff/StaffProfiles/StaffPublications/APendleton.htm of interest.
  • Totally agree - and is another reason I have hesitated for so long to publish my thoughts on this: compensation is only one piece of the puzzle and - taken in isolation - any compensation paradigm will and can be arbitraged and more importantly is only one pillar in the foundations of a healthy and vibrant business model.

    Your comments also remind me to clarify one of the points that is only obliquely referred to in my post, ie that this compensation framework is focused on large corporations. This is because it explicitly recognizes that much of the value generated by each professional is inextricably tied up in the seat they were sitting in. That is not to say the talent of the individual occupying the seat was irrelevant or marginal, on the contrary, the more 'valuable' the seat, the more important it is for the the seat to be occupied by the most talented and skilled person in order to ensure that this value was realized.

    This is why I think activities that rely less on franchise or entrenched infrastructure or relationships and more on pure human talent and ingenuity are probably better off being housed in small organizations where a more (indeed highly) variable compensation structure is workable and indeed is better aligned to the desired outcomes with less risk of distortion and arbitrage which is the usual outcome when this type of system is applied within large organizations.
blog comments powered by Disqus