Tuesday, March 13, 2012

Innovation, efficiency & the processes of industrial change

I had not intended to follow  Sunday Essay - language, social change & productivity improvement with a third post linked in some way to management, innovation and productivity, but Winton Bates drew me as he so often does. He wrote in part:

I think you are on the right track when you say that the kind of productivity change we should be seeking should be about choice. There is not much point in trying to get orchestras to play faster.

However, I think some of the problems you allude to are specific to the public sector. When you don't have market disciplines how are you going to promote efficiency without resort to things like efficiency dividends?

Winton's comment drew me because of his focus on efficiency and the linking of market disciplines to that concept. After all, I have consistently argued that current management approaches are, to use the jargon, neither efficient nor effective. Further, I have explicitly attacked the very concept of an "efficiency dividend."  These posts will give you a feel for my arguments on this second point: The Rudd Approach - Efficiency Dividends, Axe Wielding and Razor Gangs; and The effects of efficiency dividends.

As an aside before going on, my thanks to Michael Jeremy (@MikeDJeremy) for alerting me to this piece by John Quiggin, Enough of these zombie ideas: let’s be bold, in the Australian Financial Review. John and I often write from different perspectives, but I have a fair bit of sympathy for John's arguments here.

The crux of my own present argument is a simple one, summarised this wa7 in an earlier post: To my mind, Australia cannot achieve the desired productivity improvements because cost cutting is the only mechanism left in our arsenal. In writing this, I was in fact zeroing in on the prevailing focus on the concept of efficiency.

Who could challenge the idea of efficiency?  If we look at one definition of efficient we find: performing or functioning in the best possible manner with the least waste of time and effort; having and using requisite knowledge, skill, and industry; competent; capable. This leads to a definition of efficiency as: accomplishment of or ability to accomplish a job with a minimum expenditure of time and effort.

As I said, who could argue with that?

The problem I have is that efficiency itself is only one dimension of performance, and not necessarily the most important. When we focus on efficiency to the exclusion of other dimensions, we tend to lose sight of those dimensions.

The problem is partly one of measurement. We can often measure the efficiency of a particular activity or process, but we tend to do so in isolation of consequential effects.

Back in the 1990s when insurance company GIO replaced its branch structures with centralised systems intended to reduce costs, it did reduce costs. At that level, the change was efficient. Unfortunately for GIO, the company also lost market share as a consequence of the changes. Profits went down. 

As another example, the corporatisation  of electricity distribution in NSW during the 1990s combined with the imposition of performance targets certainly reduced costs. Statistically, productivity improved. By the early 2000s, the distributors were struggling to rebuild key staff that had been over-cut. The apparent gains proved illusory.

A more important problem to my mind lies in the way that an efficiency focus can blind us to the new. By its nature, efficiency centres on what we do now. How do we do this more efficiently? By contrast, innovation centres on the new.

I accept that the distinction between the two can be a slippery one.

The invention of the assembly line was a major innovation, but it was also one that centred on efficiency, that allowed goods to be produced more cheaply. It was also an innovation that directly reflected competition and market forces.

If we take the Ford case, Henry Ford wanted to sell more cars. Lower production costs opened a mass consumer market. Nor was Ford alone in doing this. Others followed in a variety of industries. Fortunes were made. So we have competition, innovation and efficiency all linked together in a major change process.

But what happened then? Once the paradigm shift, the big innovation, had occurred, the focus shifted towards improvements in the efficiency of a now established process. Innovation was replaced by improvement.

There is nothing wrong with this. It's perfectly normal and sensible. Yet the difficulty is that after a certain point the narrow focus on efficiency, on cost reduction, started to create its own problems. More and more was invested in maintaining existing systems, investments that could, at best, deliver minor incremental improvements.  The end result was another process shift that saw the assembly line replaced in part by new production techniques. However, that change was quite slow because of the sheer size of the accumulated investment in the older production techniques.

One of the remarkable changes that took place over the last decades of the twentieth century was what we can call the industrialisation of the services sector.

As had happened decades earlier in manufacturing, the previous batch production techniques common in services of all types were replaced by something akin to assembly line processes. Take the call centre as an example. Here a service activity previously carried out in a decentralised way has been replaced by a single point of contact with defined processes supported by investment in technology and in facilities. The disciplines required to make a call centre work are just the same as those holding in any manufacturing process.

One side effect of the industrialisation of services has been the opening up of previously protected service activities to market competition. We have seen this in telecommunications, in education, in law and now in retailing. The effects have been quite profound and continue.

I may seem to have come a long way from my starting point, but there is a link, at least in my mind!

As happened earlier with the assembly line, the focus in services and service delivery has shifted from innovation to process improvement. Increasingly, investment centres on the delivery of a defined range of services and activities faster and at lower cost.      

I am out of time. I will try to finish this post tonight.

Now my real point in the previous paragraph is simply this, and it is a point I have made before: the combination of technology with an overwhelming focus on cost reduction and efficiency has automated systems that don't always make sense or are not in fact the best. The consequent cost reduction has allowed them to survive when they should have been replaced. Worse, the sheer size of the investment in new systems has created real barriers to change.

What rational firm or manager would argue that the huge IT investment should actually be written off?

I now want to introduce the idea of competition and industrial efficiency.

As a general statement, I happen to agree that competition and market forces do improve economic performance. However, I have also come to wonder whether the blind adherence to market models might not have become an impediment not just to change, but to the actual working of marketplaces themselves.

Back in 2006 I ran a series of posts looking at changes in public administration over the second half of the twentieth century. Changes in Public Administration - the New Zealand Model looked at the new approaches developed in New Zealand. This was and remains the fullest application of market based approaches to the public sector. It has also had an enormous impact on Australian thinking.

As part of the New Zealand model, all Government ministries and agencies were broken into three groups depending on their customers and market positions.

  1. Contestable markets: Agencies supplying good or services to external markets for a market determined price were turned into state owned enterprises and ultimately sold.
  2. External service provision, no market: Agencies supplying services, regulation of aviation for example, remained in Government ownership but became stand-alone entities and charged for their service so as to recover costs plus a return on capital. This was meant to be fully transparent to those being charged. In practice some element of subsidisation might still be required because of externalities. In this event, the subsidy in fact represented a Government purchase from the agency.
  3. Government as customer. Where the Government was the sole purchaser, then the ministry or agency became a service provider with a single customer. In theory, this separation allowed Government to consider alternative purchases, introducing a degree of potential competition. For example, New Zealand might choose to outsource defence in whole or part to Australia, paying Australia for the service. Or buy economic advice from sources other than the New Zealand Treasury.

I was a supporter of what we might call the "pure" New Zealand model. To my mind, it provided a clear analytical structure that held out the possibility of of developing new policy approaches, including service provision. 

This view was challenged by Winton Bates in comments on this evolving post. The arguments here extend beyond the scope of this post. For that reason, I will deal with them in a later post. However, what we can say, I think, is that the bastardised version adopted in Australia was nether pure nor especially effective. There were several reasons for this.

One central problem is that the nexus between competition and improved organisational performance is actually quite hard to define. This creates difficulties for all arguments asserting any form of universal link between markets, competition and industrial efficiency.  

As a general statement, we can say that market based systems do work better then centrally planned command and control systems. We only have to look at the USSR to see this. However, this says nothing about relative performance, one market system compared to another, nor does the general statement translate easily into specific organisational aspects.

I was trying to think how to explain this in a way that would make sense, since my own thinking is still muddy.

Our views are formed by our experiences. I suppose that one of the things that caused me to challenge some current views about the role of competition in encouraging efficiency lay in my inability to establish a clear relationship at organisational level between competition and efficiency, let alone competition and innovation.

As an economist and historian, it seemed to me that both markets and competition were central to wealth creation. As a policy adviser in Government, it seemed quite clear to me that Australia's crazy tariff system had reduced both innovation and economic efficiency through the creation of an inward looking industrial structure shielded from global competitive forces. And yet, when it came to looking at performance at an organisational level, the private sector was not necessarily more efficient than the public sector; firms in sectors marked by high market competition were not necessarily more efficient or innovative than those facing less competition; while some of the most successful and innovative firms focused on cooperation with customers, suppliers and even competitors.

I suppose that we could say that we had what might be called an epidemiological problem. At a macro level, competitive market based systems delivered benefits. However, this did not mean at firm or even industry level that there was a nexus between innovation or even efficiency and the degree of competition. 

I said that the bastardised version of the New Zealand model adopted in Australia was nether pure nor especially effective.

Adoption began in NSW and then spread progressively to other jurisdictions. In all cases, the market rhetoric was the same. However, the actual form of adoption varied greatly between jurisdictions and indeed policy areas.

The "purest" and arguably most successful was the Kennett reforms in Victoria, the least effective the NSW application. In all cases, there were difficulties (to continue the epidemiological analogy) in translating broad concepts based on total populations into effective action at individual patient level.

Today, the ideas involved have been institutionalised in structures and language.

Efficiency dividends, private-public partnerships, the idea that competition is an end in itself rather than a means to an end, concepts of choice and the language of managerialism abound. The result is an inefficient mess that is quite difficult to challenge because of its institutionalisation.         


Anonymous said...

Excellent post, excellent points. Will await further thoughts with great interest.


Winton Bates said...

I agree with you that a cost-cutting view of efficiency is often too narrow. I don't think efficiency should be identified exclusively with cost cutting. At the same time, cost cutting is still an important consideration, particularly given recent increases in union power and their efforts to restore inefficient work practices.

On reflection, the model I offered of a market test of efficiency of management versus a political test is a bit too simple. Some activities (including some schools and hospitals)are supported by civil society.

It may also be important to recognize ex ante and ex post considerations and the role of luck. A project might fail to meet the market test for reasons that have nothing to do with inherent efficiency of management.

Evan said...

Hi Jim, one philosophical problem with efficiency. It assumes that the activity is entirely instrumental - not worthwhile or enjoyable in itself. I take it that this is the point of Winton's remark about orchestras.

Making work entirely instrumental might not be a good idea. There was a book called something like The Inner Game of Work - the approach to making call centre work interesting was to get people to focus on the task - things like giving people a scale to assess how much they calmed down an angry customer and so on.

Looking forward to the next post.

Winton Bates said...

Hi Evan
I am a fan of Tim Gallwey's books including 'The Inner Game of Work'.

We are having some problems with different definitions of efficiency. My concept of efficiency - which I think is fairly standard in economics although perhaps not in business studies - does not assume that activities are entirely instrumental.

In terms of individual objectives it may be perfectly efficient for a person to choose a relatively low paying job where he or she is allowed to enjoy long lunches with other staff, work at their own pace and have a great deal of say about how the work is done, rather than a higher paying job with less job satisfaction.
In terms of business objectives, a low pay- high job satisfaction culture is likely to be more appropriate in some firms/ industries than others.

Work practice become inefficient as a result of restrictions on competition. For example, it was relatively common in this country in the 1970s for restrictions on competition to allow some people to enjoy cushy jobs at the expense of other workers, consumers or taxpayers. I hope we never go back to that!

Anonymous said...

More 'efficiency':


- leaving the UK taxpayer to pick up the bill for unemployment benefits.


Augustus Winston said...

Jim - Having worked with and for the public service over a number of years (I am now self employed) I have come across this obsession with Efficiency Dividends defined as thus :

An annual reduction in the amount of resources consumed for the same level of output, achieved usually through financial and budgetary controls, improved management and administration, application of new technologies and staff cuts.

However I have my own theory on this I call it Efficiency Dividend Spill. And this is defined by me as:

The propensity for management to engage consultants and other external entities to offset the divergence of the minimum required productivity for functionality from the current output divided by the remaining resources.

Or in other words as the resources (workforce) decline to the point where departments are unable to function the slack is taken up by consultants. Although in some service areas this is not so. For some Departments there is an elasticity in service delivery.

For example I was at the RTA renewing my rego. There were two people serving behind 12 counters (10 were unmanned) with approximately twenty people waiting.

People complain about things like this but support governments who cut the public service. Well let’em complain and pocket their $2.50 tax cut and wait in line for essential services.


Jim Belshaw said...

Thanks, all. Will try to continue later tonight after tennis. For now, a few brief comments.

Winton, it is perfectly normal for efficient activities to fail a market test. Efficiency is normally defined in terms of the known; a market shift may invalidate that approach. Indeed, that is one thing that I am arguing!

Oh for a relatively low paying job with long lunches, working at own pace with high degrees of personal autonomy. Do they exist?!

Evan, all the evidence is that people want to feel that their work is of value, so as in the call centre case you have to create that value. But you also have to do it within the performance criteria set for the task. That's a challenge!

Oops. Time for tennis

Winton Bates said...

Jim, I don't understand your point.

Regarding the NZ public sector, the purchaser-provider model is about outputs rather than outcomes. Outputs equal costs, so there is no difference in practice to a system where a department is given a budget.

Rod said...

Hi Jim,

you said: "The problem is partly one of measurement. We can often measure the efficiency of a particular activity or process, but we tend to do so in isolation of consequential effects."

This is a simple statement that captures the picture very well. I always had trouble figuring out why the contemporary notion of efficiency often caused problems rather than resolve them. I think this is making it clear why there are often those problems.


Anonymous said...

Hi Jim

That's a very good point you make about the balance between investment in further efficiency as opposed to investment in innovation. There's a sort of inertia which builds, almost ensuring concentration upon further efficiency. Lose, lose.


Winton Bates said...

This discussion is of particular interest to me because I am currently writing a chapter about the concept of progress for my book.
I am attracted to Popper's concept of conditions for progress and the role of the open society and free markets in that context.

Jim Belshaw said...

Winton, the pure NZ model is very different.

Under that model ministries such as Defence become suppliers to the Government. Those ministries were expected to keep commercial accounts and make a profit on the capital invested by their shareholder, the NZ Government.

The Government was concerned with outcomes. These might take a variety of forms and combine inputs from a number of sources. Those inputs included services purchased from, say, the Treasury or Ministry of Defence.

The Government was not bound to purchase all or any services from agencies. It might chose, for example, to outsource economic advice in whole or part to another provider. In theory, this made all Government services of the traditional type fully contestable.

It didn't quite work that way. Can you guess why?

Winton Bates said...

Jim, there was never any prospect that the defence minister would purchase the services of mercenaries even if they offered better value for money than the army. The minister for finance could never literally refuse to buy the advice offered by treasury. In my view the model was fundamentally flawed.

Jim Belshaw said...

My apologies all for leaving this post and the comments too hanging in the air. It was just one of those things!

Winton, dealing just with the New Zealand model.

I don't think that it's as clear cut as that. Take defence. Political considerations, among other things, would preclude the mercenary option. But it would have been open to the NZ Government to outsource some activities by, for example, a defence mou with Australia; by use of civilian contractors; through use of other government agencies (customs etc).

The point about the model is that it provided a structured way for examining functions and services and for exploring other supply options.

Treasury is even more clear cut than defence. Perhaps as much as 90% of the Treasury function could be outsourced! The model also allowed, in theory at least, agencies to look for other income sources.

Jim Belshaw said...

Winton again, I do not know enough about Popper's conditions for progress. As you might suspect, I remain attracted to the concept of progress and to the idea that progress is facilitated by an open society and by free markets.

Winton Bates said...

Jim, I had in mind tha the kinds of considerations raised by Machiavelli would preclude the mercenary option. He suggested that mercenaries were an existential threat to government.
My general point is that there ae some core functions of government.

Coming back to your original question, my recollection is that ministers rarely had the skills needed to perform the purchaser role. They needed a departmental secretary to advise them.

Jim Belshaw said...

Machiavelli noted! I agree with you on the core functions of Government; the issue is how they might be delivered.

The minister question lies at the heart of it.

One problem with the model is that you actually have sufficient in-house skills to manage the process. A second problem is that it requires ministers and governments to articulate needs and objectives with a high degree of precision. It's not just that this is hard, but it actually runs against the democratic and political process where part of the role is to reconcile different views at the margin to make the system work.