Decentralising Responsibility: the Benefits for Mid-Caps

I believe it’s a rare case among mid-cap companies, especially those operating in developing markets, to meet one that struggles with the issue of investing surplus cash, rather than trying to balance their cash inflows and outflows.

There is a throng of mid-caps and, as with any other crowd, we tend to trust and join in the general trend – either following the latest fashion or one more traditionally prescribed by old prejudices. One such trend is that of centralising the treasury business processes, with the aims of benefiting from improving manageability through faster response time and cost savings derived from economies of scale.

Of course, business processes centralisation allows the efficient deployment of wide range of technical tools – which means IT solutions nowadays – and usually becomes a preparatory step in such deployment. Almost every chief financial officer (CFO) could recount a narrative that demonstrates the brilliant results of relative improvements in treasury function – tens of thousands of dollars (don’t forget we are talking about mid-caps) in annual cost cuttings, treasury staff headcount optimisation (also known as a reduction ) and other achievements. Nevertheless, it seems that while we’re about to win a battle ultimately it will drive us towards losing the war.

Any given treasury operation represents the final executive act of a certain resources allocation procedure. In other words, any payment should be authorised by a specific responsible manager. So in order to avoid the risk of local optimisation that usually undermines the activity of the whole system – just the same way as a selfish friend could spoil a party by drawing excessive attention to him/herself – it would be better to move deeper to the roots and focus on optimising the company’s payment authorisation rather than execution process.

The dichotomy for the nature of the payment authorisation is clear: it may be centralised or decentralised. If you choose centralisation it leads, with necessity, to the relatively strong, powerful and generally inflexible structure devoted to supporting the decision maker. Why is this so? Take the example of a company such as Velta, which has about one thousand suppliers of whom 95% are responsible for a only a relatively small portion of the company’s business needs – such as one or two types of spare part or a specific services. Thus, while being centralised the company has to maintain the ability of the single responsible individual within treasury to operate with all this stuff while keeping mistakes and discrepancies to an absolute minimum.

Please bear in mind that ‘small’ does not necessarily mean ‘unimportant’. Any manager with experience in manufacturing could recall dozens of examples where, in the absence of the timely delivery of a US$100 bearing in time, the resulting loss can be multiplied a hundredfold, due to the unexpected stoppage of the plant. To mitigate this type of risk we employ a sophisticated IT and bureaucratic system able to balance our financially acceptable inventory level against the supply schedule – the more sophisticated it is the more bureaucratic while the more interrelated the structure, the even more sophisticated it is. The circle is closed.

A Flexibility Paradox

Is anybody able to invent an acceptable sophisticated algorithm for the relevant allocation of financial resources? I believe the answer is ‘no’. The reason is clear, as one of the most important inputs for such an algorithm should be the variance characteristic of real life. So the algorithm itself should respond to the challenge of variance. But we can’t predict ‘Black Swans’! Failed with post-modern belief in ‘almighty IT’ we usually find consolation in traditional bureaucracy replacing machines with the group of treasury staff who make calls, calculations and monitor the business units.

It’s an interesting paradox. When starting with the attempt to address variance in the reality of business life by centralisation we inevitably end up with the relatively rigid set of tools and procedures we know well as the ‘budget’ or, more specifically, ‘fix annual budget’. We’re simply trying to become more flexible in respond to day-to-day challenges by becoming less flexible!

What about testing the decentralised solution? Classify potential types of payments, concentrate authority for initiating the most important (which are not necessarily the largest ones) in the head office and spread responsibility for all other payments across the various business units. Probably this approach would require more personnel throughout the whole company and several different – even incompatible at the outset – IT systems, but the payback would be gaining more flexibility in return. Business unit heads are much closer to the problems such as a temporary lack of bearings and in-time maintenance of the equipment than head office personnel. They also have clearer understanding of grass roots risks, their consequences and ways to avoid them.

I have successful experience of employing this kind of decentralisation in a mid-cap mining company developing three geographically spread out open-pit mines.

Sales were partly decentralised, with head office responsible for around half of total sales and individual business units for the remainder. As a first step we agreed head office would retain responsibility for a handful of payments, namely those going to energy suppliers for power, bank interest settlement and taxes. We estimated that those three payments accounted for 70% of the total cash cost, so we agreed that half of the income collected from business units’ own sales should be transferred to the head office account and be used to cover these responsibilities and the rest could be used locally by units at their discretion.

It took three months to tune up the system. During this period we faced a strike at one of the units, caused by the inappropriate spending of funds by the local plant manager. Head office didn’t need to help because we proclaimed a ‘zero tolerance’ policy for wrong decisions; thus motivating local managers to take responsibility seriously and showing them the limits for mistakes. In just a single financial quarter the overall system had been healed. Our approach was able to address natural minor fluctuations in cash flows and gave us an algorithm for re-estimation in case of substantial changes, either in the external business environment or in the business model.

We allowed our business unit managers to become true decision makers, build trust within the company and free up head office personnel’s’ time – both among top layer and mid-level staff – for more engagement in analytics and strategy issues. Finally, the most valuable result we achieved was that we allowed business units to obtain the resources they needed upon request in time – and not through tightly-prescribed schedules in most cases irrelevant to the real circumstances.