Returns on Human Asset Investment Dr Keith Bradley, Executive Director, Business Performance Group, London School of Economics

Lord Butterworth

The Business Performance Group at the London School of Economics is an interdisciplinary management research institute specialising in business performance. When we met in October, Keith Bradley gave us a fascinating address on business performance and human assets. We now look forward to being brought up to date on more recent developments.

Keith Bradley, London School of Economics

I think I will do precisely that. I am going to rehearse a little where we ended the last session, then develop it a little, and then open it up for discussion. At the end of the discussion, what we hope to get is to see whether we can formulate something which we can report regularly in the form of return of people assets.

This is not a panacea; it is the terrain, rather than a road map. What I do hear, on my kind of stamping ground, is people increasingly asking for a understanding of the returns on human assets. I think what is driving that is that sustained competitiveness in the future is going to be derived from human assets. This will occur more as companies move their values to human assets, and recognise that the value of the company is invested in the people, the networks, and the know-how. I think that where we are arriving at is that accounting does an extremely good job with physical assets and financial assets. Even accountants would accept that there are some concerns about understanding or reporting non-tangible assets. So what we are trying to do is to move the agenda forward by finding a way in which we can report on these.

Now there are two things to bear in mind here. One is that we are not trying to do what was done in the 1970s. That was to try and shoehorn the human assets into the balance sheet. That is something that we are keeping well away from. The other thing is that we are in no way raising any kind of concerns or criticisms of accounting numbers. We are not in that kind of paradigm or area; we think that accountants and finance directors do an extremely good job. Whilst there are plenty of criticisms being kicked around, what we find in our group is that, in terms of actually understanding physical assets and financial assets, there isn't much better practice available than what is going on at the present.

What we do think is needed, and in some firms we work quite closely with finance directors, is a way of finding something which is numerate, which augments and complements the financial indices. We need something which can be reported internally and eventually maybe reported in the annual reports.

The first discussion, which we had last time, concentrated on a very straightforward thing: can we value human assets and, if so, how can we do it? How can we do it in a way which provides us with indices, which can be reported regularly? These indices can be used for strategic thinking, so that the human resource aspects of companies can be brought into strategic planning and strategic thinking. This is just summarising what was said last time.

To mention some of the uses of human asset value, you can talk about the cost of staff turnover, the cost of redundancies, the turnover of management. So how do we measure this? Those who are familiar with accounting practices will find this is really quite straightforward, quite elementary. We are simply doing this on the basis of trying to estimate the cost of replacing an employee, in the same way as accountants would actually go about looking at the cost of replacing a physical asset. Now this in some ways restricts our discussion. Some people would say that perhaps it restricts it to the lower end of organisations, and to organisations in which you have many jobs which are the same, rather than to what I would call the prima-donna end. There is a lot of work going on at the moment, trying to find out the value-added of a special individual in your organisation. For instance, a senior partner who brings in 80% of the billings; this guy is so important, how can you somehow value him? We are still grappling with that, but in the end, what happens is that we end up with boards full of mathematics. Usually, when you present the results, there are one or two people present who can think of a variable which you haven't thought of, and which is the most significant variable. So I don't want to focus on this aspect so much today.

We are encouraged, by some of our clients and some of the people we talk to, to think that there is sufficient reason to try and organise this valuation, from about the middle-management level down. They think that in many organisations it can be of huge value. We can organise that type of thing, and that's where we are doing it.

Let me tell you again the genesis of this idea. It started in 1984 when I was at a board meeting of a large engineering company in the United States. This was a company of which it was said that its future was in its ability to recruit the brightest and the best young engineering graduates. Worldwide they were recruiting 2,000 engineering graduates each year. The line going through the board, a line from the human resources director, simply said that we had reached our target of 2,000 and we had, within that total, an annual turnover of 15%. They thought this was probably better than their competitors. It was a slap on the chest, we were doing very well, and we were going to move on to the next item - having occupied 15 seconds of the board meeting.

So I said, "Can we just ask a few questions, because these people are important to your future? You are an engineering company, taking the brightest and the best from the top universities in the United States. How much are you investing in people each year, on a simple thing such as training?". Then you know the sort of thing that happens, people start quoting telephone numbers, but let's say the consensus was about $20,000-$25,000; around £10,000 at that time. And then the second question I asked was, "Well, how long are these people staying in the company before they leave?" We settled on three and a half years. So, before they leave, the sunk cost is about £35,000 per person, with 15% turnover of 2,000 is 300, and 300 graduates leaving each year multiplied by £35,000 brings you to about £10m.

This was a kind of leakage from the organisation, an incipient leakage. They had already established that this recruitment was an important part of their future, and yet before those questions were asked, they were congratulating themselves, and moving on to the next item.

This started me thinking, £10m was about the cost of the Chairman's jet, and if somebody had nicked that jet , that there would be blood on the walls. So I started thinking along these lines: was there a simple, Mickey Mouse way in which we could retrieve information, order it, and regularly report these kind of sums? What we had stumbled on was actually trying to value the human assets, as you would value a physical asset, via its replacement cost.

It is a good system because you can do it very straightforwardly, and you can depreciate the asset, and you can see how it is depreciating. So, what we have is two very simple equations, and the method that accompanies them. This is what we were talking about last time; it is a way first of valuing the replacement cost of an individual, and then of multiplying that cost by the number of people who are in a particular function, business unit, or level. You will see, as we go on, that the value of this system is, in fact, its flexibility. It's a way of finding the value, and once we do that, we can see how the asset is depreciating. Because once we know the value of the employees in a particular function, and we know how many people are leaving from that function, you can see that it is possible to depreciate and see how much the asset is depreciating. So, that's the very straightforward approach we use.

Elements of Employee Replacement Cost

Now, how we do it in practice? When you talk to companies, there are a lot of people who concentrate on salary costs. We don't think this is centrally important; it shouldn't actually come into one's calculations. The reason is that for a production engineer, for example, it would be the same both inside and the outside the company. It is not a value-added figure which companies have created, it is a constant. What is not a constant is the sums which the company pays hiring that person: its recruitment costs. So the first item, the first investment which worries me, and the first element of the human asset value, is in fact the recruitment cost. We add to that the training, the formal training cost, and we add to that the informal training cost. You will see in a moment what we mean by that, but basically we mean the opportunity cost of, perhaps, a department head being taken away from providing services, or producing products, and spending his time doing on-the-job training.

We add to that what we call the inefficiency curve, the amount of time it takes for someone to come up to 80% efficiency. That is an investment which companies have made. Then you get a value, the valuation of one person. Once you have found out the employee replacement cost, you can multiply that by the number of people in the function. You can get a value for the full workforce, or a particular level of the workforce, or a value per business unit; you can cut the cake whichever way you like.

Human Asset Value: Variance by Cost Item

Very briefly I would like to show you, in a little more detail, how we do it. You can see that the first two elements, the recruitment costs and the formal training costs are usually very easy to find. The information is there, the allocation of the recruitment budget over the number of people you have recruited. But you must take into consideration interviewing time, reading the papers; people often underestimate the cost to management of this. Secondly, you include another thing which is easily available; formal training - that is, not only the internal courses, but the external courses and the allocation of that budget over the number of people.

The other two are perhaps more intangible. Informal training is an estimate of the opportunity costs which managers and others pay in off-the-job/on-the-job training. In companies which have a high level of company-specific information, it is often quite an important added cost, and one which is often not taken into consideration. The last item is the time it takes for someone to come up to, and be running at, 80% efficiency. And that, for those of you who were here last time, or for those of you who have seen it documented on our database on Lotus Notes, is it.

To expand on 'opportunity costs'. What I mean here is that frequently, in situations where someone needs on-the-job training, he is trained by his supervisor, or by the department head. That means that the head of department has to come away from what he is doing, from adding immediate value to the company. Instead he has to spend the time sitting with the trainee, going through procedures. Now that is a cost. And of course the person who is being trained is not adding value either.

Those are significant costs, and you will understand, for example, in situations where companies have high levels of company-specific information, that can be quite a considerable cost. What we encourage people to do is to think in terms of knowledge. There are two axes to it, one is the cultural knowledge and the other is the technical knowledge.

What you have here, if you join it up in a terminal matrix, at the bottom you have industry-specific knowledge, and at the top right you have company-specific knowledge, and this is on both the cultural axis and the technical axis. The cultural axis covers the way things are done, the informal systems. If you consider those people who have high levels of investment on the cultural company-specific axis, and high levels of investment on the technical company-specific axis, these are the people who you can't afford to lose. Whereas, you might say that people lower down the chart who have industry-specific knowledge might be treated differently.

Let me tell you how this works out in a practical way. In many companies. managers are all treated the same. If you are a manager, that means you that when get to a particular point you are given the same remuneration, the same car, everything similar. What we did was a kind of body-scan of information, a body-scan of the managerial ranks of the organisation, and we found that the engineering managers and the technical managers were in one category, the warehouse managers and the transport managers were in another category. We felt that maybe the company should re-think treating them all the same. The cost of replacing one group of these people would be significantly higher, judged by our kind of evaluation of them, than the cost of replacing the other group of people.

What we have then, in terms of the concept of human asset value, is a new type of management information. "New" is probably overstating the point. The information is there, but I think that it needs to be organised, and organised in a way that is easily retrievable, and can be reported on a regular basis. The beauty of something like this is that it is very cost-effective. It doesn't need smart people to be constantly involved in it. Once it is set up, it can be retrieved, and this type of information can be sensitive to the technology of the firm - if the firm is computerised this can be a computer system. If it's not computerised, it can be on filing cards and so forth.

It is a flexible system; it can be cut in terms of whichever function is needed. Of course, predicated on that is the management action, because it is a body-scan of the organisation. It identifies the trouble spots, identifies where you are haemorrhaging, or if you are not haemorrhaging. The idea of it is that it provides a strategic tool which you can use in tandem with your resources. If you take resources like training, which is a high expenditure item in some companies, it is often splashed around like a kind of public good: everybody should have it - it is a good thing. But this approach of ours is moving much more towards a kind of key-hole surgery. Once we have identified the points, we can home in on them; we can provide some effective training or management input. And because we are reporting at intervals, it can tell you how effective you are on that particular issue.

With these indices you can have sessions with all your department heads, or your functional heads, and you can sit down and you can arrive at an optimum level of human assets value, or the depreciation of that asset. The management system should be oriented towards arriving, or keeping, or maintaining, that optimal level. It is not axiomatic that you will want to minimise staff turnover; in some instances you may wish to have a healthy turnover, but in the top right hand quartile you would not. It is actually managing the system to those optimal levels that we encourage.

It is because the system is relatively accurate that we do encourage companies to arrange for these kinds of numbers to be related to partial remuneration, to a bonus system. Therefore you would not only have a production bonus or a profit bonus, but also a human asset bonus. We say that this is results-oriented. In the end I will just show you how some of these things tally up.

Human Asset Value of Supervisors in Three Companies

This is an accounts function, the human asset value of supervisors in a banking firm. Note that they are dealing with the accounts, and you can see the total supervisory human assets in this tiny little group is about a quarter of a million pounds, about £16,000 per supervisor. Because you have an annual turnover of five of these people you are losing, in that tiny function, that very small group of people, about £80,000 per year. The good thing here is that you can actually relate it to trying to reduce your turnover, to see what effect that has on your operating profit margins.

Let's concentrate a little bit on customers because this is important. Let us assume for the moment that we know how to value customers. Now what seems to be talked about with customers is the following kind of model which shows a lot of the 1980s type of discussion. You have your suppliers, you have your business, and you have your customers. There was a lot of contemplation around this particular focus, that it was the external customers that you had to be very sensitive to. A lot of strategies and a lot of thought processes went into satisfying those customers. Now very quickly we have moved on to think, and this is nothing new to anybody here, that if you blow up the business, what you see is that the business can be conceived of as a series of supplier/customer chains, and here we see the design group as a supplier of the packaging, and the packaging, as the supplier to marketing, and so forth. So you get these horizontal chains flowing throughout firms, or throughout organisations and supplier/customer chains.

The next thing to bear in mind is that this is, again, nothing new. What we find is that some of these supplier/customer chains are more effective than others. What we are looking at is the transaction costs between each supplier and each customer. Here is where the transaction costs are pushing into your profit margins, given the kind of competitive market where the market establishes price. As a business, part of what we are concerned about is trying to understand transactional costs, because we are trying to reduce them. If you are trying to reduce transactional costs you have, first of all, to understand them, and you've got to have a way of reporting them regularly, to see how well you are doing. What you have here as the next slide is a way of understanding that.

In that kind of model, the customer/supplier chain not only runs horizontally through an organisation but also runs vertically as well; the corporate centre is often the supplier of financial resources, and sometimes personnel resources. It is important to understand the vertical chains as well as the horizontal chains. We want to end up with something which is straightforward and simple, and a way of reporting regularly an index which actually measures the transaction cost.

How we do that? It is in fact fairly straightforward. The first thing we do is lay out for an organisation the supplier/customer chains. I cannot emphasise that sufficiently. I am sure you go into organisations as I do, and they provide you with an organigram of how things work. They say "this is the supplier/customer chain" but that's the official system; the actual system is very different. Actually working that out can lead to major gains in performance and efficiency in an organisation. The first thing is to map out the real picture of what is actually happening between the supplier and the customer.

The next thing is to spend time with the customer, and what happens is that we brainstorm and we raise certain criteria of performance. When you are thinking of your suppliers, would you order your criteria of performance; things like price, reliability, lack of bureaucracy, modernity of goods, responsiveness to ideas? These ideas don't necessarily come up every time, but they do come up enough to matter.

The next thing we do, if we are in a business unit or a firm, is that we harmonise these particular criteria of performance throughout the business unit. With any business unit, every customer group has prioritised the top ten criteria of performance. The next step is very straightforward indeed, asking the customer groups to weight these criteria, to tell us how important these criteria are by putting a weight on them. Once that is done, the ground work is over. What happens then is that we install, at regular intervals, satisfaction audits, based on these criteria of performance.

Weights assigned to Performance Criteria for Service A

How satisfied are you with price? How satisfied are you with the lack of bureaucracy? Then we get a satisfaction rating. The beauty of this is that we are able to compute the weighting with the satisfaction scale, but we can also weight the customer because you have some preferred customers. We have dealt with a bank recently, where something like 8% of its customers provided 80% of its business. Clearly you would want to weight them in a much higher way that you would be weighting a smaller customer. This gives you a performance index, either for each measure or you can aggregate these up, which shows the transaction cost between the supplier and the customer.

The good thing about this is that it provides the supplier with a road map of how to improve his service, or how to improve his product. It tells him very clearly where to allocate his resources. Frequently we find that suppliers are playing this game blind, because there is no interaction. They might be spending their resources on reducing bureaucracy, when their customers don't rate that problem very highly.

Satisfaction with Service A according to Performance Criteria

The other thing which we find is that customers and firms that are using an employee audit with only a satisfaction score should have a health warning on it, because it really can provide you with very misleading information. This is mainly because you can run away with the idea that 75% means we are all doing extremely well, and therefore we don't have to worry about pricing. But the very fact that the customer rates the price extremely highly means that you should still be allocating some kind of resource to the issue. This applies either to the internal or external customer: it doesn't really matter which. What does matter is the emphasis in the 1980s which was placed exclusively on external customers. I think this kind of fetishism has actually distorted the reality of improving efficiency and performance in a corporation. If one can reduce the transaction cost in a small way, it is probably better and more effective than doing the Big Bang with the customer at the end.

We have this broken down into the functions; the accounts, the cashiers, the reconciliations section and then we also have every activity in the back: the requisitions, customer service, interest-bearing accounts, receipts, investments, and so forth. We also have the interface with external customers. What we find is for the aggregate is that best practice is with the requisitions. The beauty of this is that it has the ability to operate relatively over departments, or over activities, so you can home in on best practice and you can see that as a strategic objective, bringing other activities up to best practice.

Also we have analysed the customers, broken down into small, medium and large, and there are slight differences in terms of how different sized customers rate. Interestingly, you sometimes find that large customers are not too worried about bureaucracy, because they themselves have it. Large customers are more sensitive to other areas. Small customers, who are not used to bureaucracy, do mind. That provides you with an index which you can aggregate up so that overall there is a rating. The other advantage of this is that it can be reported over time, and that gives you another comparison. So you have got two comparators: one comparator over activity or function within the organisation, and another comparator over time.

How often would you change the criteria?

The model which we have got allows for changing of criteria. It frequently depends on the nature of the business; some businesses have criteria which don't change very often, others are much more sensitive to change. In some engineering companies, where the product or service is changing very rapidly, and the lead times are shrinking on a weekly basis, then the criteria in our model will take those things into account.

When you are working internally, does this tend to lead to passing the blame backwards to the supplier all the time, rather than the supplier being able to imply that the customer is, perhaps, at fault in the way he is using the product?

I think that's right. I would probably put it in more muted terms than that, because this whole activity is our product at the LSE, but you are absolutely right. It is unashamedly looking at the supplier from the customer's vantage point. It is the customer you are really trying to please here. This is not a panacea; it's a technique which can be tweaked, and it's a terrain rather than a road map. One of the things we have found is that a lot of internal suppliers are allocating company resources to areas where there is no real need for them.

One criterion which I find surprisingly absent is that of time.

I have not put everything up on the charts. Doing things in a timely way is often a big thing. What we have here is a human asset value. We can see that the investment that firms are making in their human assets is an investment we know how to quantify. What I have just shown you is what I would encourage you to think about as a quasi-benefit. What you have is an internal customer audit which is a quasi-benefit and what we have been doing is putting those two techniques together and reporting the ratio.

When you spoke about human assets depreciating: was that the assets walking out the door, or the asset that remains?

There are two things; one is the value of the asset which remains, the second part is the fact that you know how many people are leaving. If you have properly assessed the asset, then you can multiply it up and assess the depreciation. Then you have a very good idea of the leakage, the haemorrhaging.

You mentioned that this is, in a way, off-balance sheet; that you are not intending to change the accounting numbers, you are using financial-type numbers to focus on issues. But is there a sense in which you are also saying, look, if we did things otherwise in this company, the company would save £150,000? This is not just speculative financial information; you are saying that there is a decision to be made.

You can see that numbers of this kind can provide a very good way, of sitting down with a department head in the accounts function and saying "Can you really afford this? You are losing £150,000 a year, can you really afford that? If you can't, then what you should be thinking about is targeting some of the resources, training or backup, to this particular group of supervisors to try and reduce the turnover." This, then, becomes a benchmark and it also becomes a target in that next year, we don't want to lose more than three people. If these people are crucial to the company, we will try and operate over a timeframe of keeping this group happy.

Do you see it as a ranking which lets you decide where you are going to get values on your human resource development spend, where to focus the scare resource?

I think that is right. I think the beauty of it is working in tandem with these spending departments in companies, and rather than a scatter-gun which might kill somebody, this is a more determinedly focused activity. Because it is actually being reported at intervals, you can see just how effective you have been in bringing different functions in firms together. Frequently you find that the human resource function doesn't talk to the training function, which doesn't talk to the work production function. This process provides a way of understanding that this is everybody's responsibility; we all have a role to play, we all contribute our parts here. This isn't a way of attributing blame, it's a way of measuring how effective we have been, and reassessing the effectiveness of it, it's that type of tool.

One of the things I am not happy about is that, as an accountant, it would lead me to think that I should be trying to minimise that number. That may not , in fact, be what is best for the business, because the business may be enlivened and invigorated by staff turnover. We would die if we had zero staff turnover.

You must be an insurance company. The extraordinary thing about insurance companies is that they really have a problem because they don't have staff turnover. Because of the history of their oligopolistic control of various markets - opening the door at nine in the morning and closing it at five in the afternoon - people had lifetime employment. You give them great benefits, but times are changing, and the insurance companies I deal with are saying, "How do we create some kind of turnover to bring in new ideas". This is why this index is not necessarily good if it is on the positive side. It is for the strategy groups, the managers, to think what is the optimal level of turnover. In a particular situation you might wish to have some turnover, but if that is your desired goal, what we say is that it is very important for you to understand the cost of that turnover, so that you can actually manage the cost of it. Maybe you might not be so liberal in your spending on certain items if, for example, it was your desired goal to create, within the division, a certain level of turnover. What this does is allow you to have a very good understanding of the financial implications of some of your strategies on human assets.

You very emphatically distanced yourself from putting these values on the balance sheet, as people in the '70s tried to do. Is that because it comes into the 'too difficult' category? It does address the point that I think the last questioner was touching upon. You may have no staff loss at all, but still have a depreciating asset, and your methodology doesn't address that. At least, as you've told us so far, it just addresses the cost of staff turnover, rather than the value of those remaining. Have I understood that correctly?

No, because, if you go back, we can find out the investments in human assets, without any turnover whatsoever. Let's say, for the sake of argument, that within a particular function we have ten people, then the asset value might be £100,000.

It doesn't attack knowledge decay; the depreciation is known, but not whether the training is appropriate or sufficient.

We have some other indices which we use, which get at that point. This is a layered process, and this is what we think is one of the basic things which you install. Then you can develop other things on top. You are absolutely right; it doesn't take into consideration the knowledge decay and so it doesn't depreciate the asset value in those terms.

The depreciation rate you are using is the staff turnover rate. We are almost missing one side of the equation, the new ideas side, and maybe the human depreciation should be the time for which that person adds more value that someone else.

There is an opportunity value as opposed to an opportunity cost. Opportunity value flows from turnover because it leavens an organisation, it stimulates new ideas, it challenges the perceived wisdoms and it is absolutely critical. If you stop swimming you sink. The technique does not address that part of the organisation which is going to be significantly enhanced, or otherwise, by the ideas; it stops short of that. It breaks down in the area of that person who does make a significant difference to the organisation. Where it works very well is dealing with something like Safeways, where they are losing £23m a year with the turnover of the people on their checkouts. It really addresses parts of organisations where you have a function which is mostly straightforward and repetitive; it does not address the partners in a law firm.

Can I just pursue that point? When we had our new measures of success seminar, we had a presentation from a guy called Alan Benjamin, chairman of a software company. His starting point was, "This is a knowledge intensive business; what I have got to try and measure is what I call the knowledge-bank." I am not the best person to describe this but I think Alan Benjamin would be very interested to come on some future occasion and talk about it. This was an attempt to try and deal with the high level knowledge worker, the innovation contributor, whereas I think you are saying that your process works very much better on the repetitive side.

Well a lot of places where we have it installed are knowledge-based companies. I think there is a lot of hoo-ha about knowledge-based companies. Talk to my students, who are seduced into joining these knowledge-based companies on the basis that they are going to have great responsibility, that they will be given great openings, bright ideas are going to be immediately enhanced, etc etc. It's a lot of crap. As far as they are concerned, what they find is they are doing very repetitive jobs. In the knowledge-based companies there are a lot of company-specific processes, a lot of company-specific techniques, and a lot of company-specific ways of servicing customers. These have to be passed on, and that is a huge investment; an investment which is often underestimated by knowledge-based people. There are people, but I think these are unusual people, who actually do make massive differences to companies, because of their genius, or because of their lateral thinking nature.

However, our feeling on knowledge-based companies is based on putting systems in place. We are dealing at the moment with a large advertising agency. The danger we find there is that we see that there are few current creative geniuses in the company. Most of the company is a fairly organised process. The problem is that there is a kind of contamination effect from the top; everybody wants to be a creative genius, and the whole thing is just chaos. I feel that while this deals with the kind of repetitive thing, I think we can sometimes overestimate what knowledge-based companies are all about.

Could I make a comment on that? I understand the cultural knowledge end but I guess there are three other aspects of that. One is related to your customer base, which may actually be specific, which is not part of the culture. The second is in terms of the supplier end and the third is in terms of any strategic alliance, or other relationships. There are the critical relationships, and some intrinsic value that they bring.

You then begin to do a value ranking of your human asset, which is not necessarily by purely financial criteria, but as a management judgement of the business contribution. What is the difficulty of replacement of that person and how is that person able to facilitate change in the business, or able to share knowledge? And that is balanced against the thing that I guess others have addressed, which is that you need external experience because that is often the agent of change that enables you to move forward faster. Therefore it is almost that, if you get the ranking right, you can afford to spin off at the bottom and engage new, fresh skills and mindset. However, there are the people at the top who may have cultural knowledge and relationship knowledge that is critical, and once you have identified them you can always plan against a background of understanding the cost in a more substantive way.

I think that's a fair point. My feeling is that what sustains the people at the top, from one brilliant idea to the next brilliant idea, are a lot of people doing some very good process work down below, and that, I think, should not be underestimated. That is my own theory and what you have done is nicely brought into this the idea of customers.