Following from Ken’s post the other day I spent some time in idle thought. For the moment I’ll disregard my problems with aggregate productivity statistics (many of which are covered in this Grattan Inst paper). I’ll also disregard my feeling that ultimately productivity growth is beyond individual government policies and wealthy countries like Australia are sticking close to a frontier that is determined by the global stock of human knowledge – a frontier that is receding at it’s own steady pace. When speculating on this I got caught up on two issues which could use their own brainstorming.
Corporate Governance
It is far too tempting to reduce productivity to labor productivity, and then to reduce that to the effort and ability of workers. Whilst economists tend to avoid this fallacy, broader discussion, particularly in the media, tends not to. Pundits who call for labor reforms that would make it easier to sack bludging workers, or headline writers who translate a call for greater productivity into a call for people to “work harder” or work on their human capital all into this trap. But it is obvious that the output of any worker is constrained by where they work. All the individual effort in the world won’t make an ageing machine work faster, or speed inputs through poorly constructed processes and documents through labyrinthine corporate bureaucracy. Innovation and technology are the base of all productivity growth, but they don’t show up in a worker’s productivity unless the powers that be allow them to use it.
Consider the “cold shower” effects that were an unexpected bonus of trade liberalisation. Where the conventional argument for low trade barriers merely thought that it would allow resources to flow to more productive sectors, the cold showers indicated improved productivity within industries that were newly exposed. The continues to be observed despite a paucity of behavioural models to explain it. I think it is sufficient however to merely relax the assumption of maximising firms.
Lets say that firms have a target threshold of performance. Should they get more than that, that’s a bonus, but not terribly important. It is more important to avoid falling beneath it. A firm at this threshold is content to leave fruit unpicked, avoid the innovation and restructuring and improved equipment etc. that would improve productivity even if it would be fairly easy to do so. If we assume non trivial barriers to entry (such as the costs of establishing a business) that are minimising pressure from domestic competition, it is only when exposure to trade puts the pressure on that all these options begin to look a great deal more appealing. Only then do firms begin to pick the productivity fruit that was there before liberalisation, but which they neglected. This kind of argument is much more easily found in labor economics 1 literature than trade literature where it is used to explain occasional increases in productivity when unions manage to transfer rents to labour, or the difficulty of empirically demonstrating the seemingly obvious relationship between minimum wage increases and unemployment. Here the squeeze comes from labour costs, but it is still management being shocked into doing what maximising firms would have already done. 2
This is intuitively true – any experience in a large company leaves you thinking that processes that could be easily improved or wondering why you must reboot an ancient computer twice a day for want of $40 of RAM. Importantly it suggests that there are structural impediments to productivity growth even whilst potential productivity is increasing. How do we get firms to improve productivity as it become available. We can’t infinitely liberalise trade, we probably don’t want to wait until foreign firms force our management to act and we would be ill advised to choose firms to shock into action with mandated wage increases.
First up, if it is firm wide satisficing in the Herbert Simon sense, we may not want to address it. Here management is minimising the risk of company failure at the expense of maximisation. If we value stability over income we might consider a tradeoff worth making. On the other hand we might also develop a greater welfare/social security system and target lower frictional unemployment as a way of protecting people from this risk whilst allowing optimal deployment of resources.
However I am inclined to think it is more related to agency problems (including shirking and bad incentives for management) and informational problems.
With agency problems the management is minimising their own risk of getting fired or disciplined. The threshold they aim for is merely what it takes to avoid their principals (higher management, the board and owners) taking action against them. This threshold would likely be related to wider benchmarks, such as the average return on capital amongst listed firms or relative share price performance – lacking inside information, these wider benchmarks may be the only way outside principals can gauge how well management is doing and a firm that is underperforming in regard to potential may be doing fine compared to others giving no call to action. The threshold would also be related to the difficulty of taking this action. A management and board protected by elaborate election procedures and two tier share systems, AGMs held in inaccessible places and contracts that would scorch the earth if they were removed would face a far lower threshold. The cost of moving against them would outweigh the benefit of better management.
As long as they are above the threshold, management thus is free to shirk and avoid the effort they’d need to put into adapting innovations and changing processes and expend it on things like long lunch interviews with the AFR. They are also free to divert resources away from productivity friendly but unsexy places (like having computers that don’t crash) towards those that magnify their own prestige but with little return for the company – CBD office locations are one, as are marketing campaigns that heighten the profile of the company and thus its management but fail to bring in new business 3.
The pundits that fervently desire labour market reform on the basis that the threat of dismissal is all that keeps most workers from sloth, but do not appear to even consider applying the logic to management. This is worth investigating however.
So what could be some responses. Incentive contracts have been tried, but they seem largely to have created a wider gulf between remuneration and performance that was not there beforehand. Alternately there could be regulatory steps in order to make discipline of shareholder democracy and activism work closer to the way it’s meant to. This would involve making elections more transparent (allowing internet voting), outlawing poison pills, making executive pay determined by shareholders rather than a board of cronies, etc. This kind of regulatory approach shouldn’t be unpalatable to liberal types since it merely exposes management to the market discipline they advocate for workers. The lack of evidence that management performance is related to remuneration suggests this discipline is sorely lacking. This might be one way to get management to continually adopt and innovate rather than waiting for them to be shocked into action.
This would only attack part of the problem – the informational asymmetries are still present. Mandated and accurate statements would only allow outside principals such as shareholders to recognise and take action when a firm fails against a wider benchmark, as discussed before. Potential improvements within the company may need insider knowledge to be recognised and management may still shirk from these if they require effort and the firm is performing well compared to the benchmark. Worse however is that an industrious management may be punished. Work towards thousands of tiny improvements that involve both a great deal of effort and tacit knowledge (as opposed to tangible and flashy capital investments or layoffs) or engage in non sexy R&D may not be apparent to outsiders until after management has been jettisoned for failing against the benchmark. I think this is the bigger problem. Executives are well remunerated but hardly secure in their tenure so it is less likely to be that they are shielded from incentives than that these incentives do not relate to potential improvements in productivity.
Charles Lindblom once said that firms were islands of command in a market sea. Where in the market tacit knowledge is encapsulated in price signals allowing co-ordination, in a firm they are embedded in institutional knowledge. It’s hard enough for management to discover what productivity improvements are possible in their organisation. It is next to impossible for outsiders, including many owners to discover. If it was, it wouldn’t take shocks and cold showers to force managers into looking for them and principals to start prodding them. Potentially productivity improvements would be more continuous if this insider knowledge was better shared. Possibly even this would
But how? There are all sorts of steps management can take to improve their own understanding and airport bookshops are full of inane tips. But if they require effort and this just runs up against the management shirking problem. How could corporate governance be changed to diminish the insider-outsider information problem? I don’t really know. The only thing that comes to mind would be encouraging employee representation on the board. Complaints about little everyday productivity sapping things like outdated equipment or unresponsive management structures could then be more apparent to other board members at least. They could then prod executive management into exerting the effort into dealing with unsexy problems. Possibly further stakeholder representation (such as bank lenders on boards) who have a longer horizons and are prepared to wait on R&D and restructuring.
These present their own problems however. What other ideas are out there?
Monopsony and Human Capital.
Around about this time many school leavers and university graduates are looking for work. They’ll also hear about a skills shortage. They may be bemused. If there is a shortage of IT specialists and programmers, or accountants, that requries a skilled migration program why are they finding it so difficult with their computer science and accounting degrees? If there is a shortage of tradespeople, why are apprenticeships so crappily paid. Why is every employer looking for experience and none are giving it. They’ve just spent 3-4 years studying courses that had the interesting and stimulating parts stripped out in favour of annoying group assessments and other tasks that would make them more “job ready” – only then to be deemed not ready for a job.
Human capital is oft cited as a way of making more productive workers, but employers seem to be demonstrating that on the job training and experience is an essential part of this and one that can not be substituted for by redesigned, “job ready” undergraduate education. Yet they are loathe to invest in this when employees are able to move from company to company, taking their human capital with them. Why put in the effort and money when someone else can poach them? Unless you expect to have a degree of monopsony (buying) power or expect a great deal of loyalty, there’s no point.
Workplace flexibility cuts both ways and a flexible workplace expects loyalty from neither employers or employees – it would be strange to call for it after 20 years of blaming Japanese economic woes on a lifetime employment model. Do employers want to be stuck with mistaken hires and do employees want to be tied to one company, or have their lifetime career determined by graduate recruiters? Neither employers or workers may be keen on tying their hands with very long term contracts or regulation that makes it difficult to both quit and resign. If we want to retain a system where employees change employers and employers change employees as circumstances require, we need to think of other ways to foster this human capital.
What to do here? We can’t make the workers pay for it even if they benefit because the nature of the training precludes it. We can make people bear (part of) the costs of their university study, but, as above, such study is deemed not a substitute for experience. Would the potential workers pay for their experience with free labour? This is already done in traditional apprenticeships, but we still have complaints about a shortage. This has also been adopted in the form of unpaid internships in the United States. They lend themselves to exploitation, but I also think they inordinately benefit the well connected creating an inequality of opportunity that extends through generations.
Another option is to maintain a strategic reserve of unemployed in the economy to keep wages down, increase monopsony power and allow firms to appropriate the return on human capital they invested in by paying lower wages. This doesn’t work on several levels. It belies the entire point of increasing productivity (higher living standards) by forcing unemployment on the public, as well as inviting potential unrest. The unemployed will lack, or lose, the on the job experience desired and they won’t serve their purpose as potential job rivals. It also doesn’t stop poaching between individual firms. Some of this can be avoided by having this strategic reserve overseas. In boom times temporary skilled workers are imported to keep down wage growth, but there is no avenue provided for citizenship and if unemployment rises they can be deported. This seems to be in operation in Switzerland. It also seems to be ethically abhorent (almost literally creating second class citizens) and again does not stop poaching between firms. Whilst a potential approach, particularly under a modified 457 visa scheme, I would dismiss it outright.
Wage subsidies have have been trialled elsewhere as a way of helping the long term unemployed with disappointing results, but they might be made to work. It is very difficult for the government to accurately determine whether human capital is being developed instead of firms exploiting a way to have subsidised low skilled labour – or whether they are merely imparting firm specific skills that they would have done anyway.
Given I still favour a system that allows employees to move around and I dislike employer monopsony, I can’t really think of a way to foster this “on the job” human capital more than is being done so.
1 I suspect because it’s harder to start with industry wide analysis, Cobb-Douglas and continue with a universe in which there are no firms.
2 This produces a hypothesis I’m not going to go into but is interesting. If people are concerned about wage growth relative to measured productivity, maybe a tight labour market is what is needed to get management to get their act together. Wage growth now to force investments in productivity.
3 If you rig the voting system enough, you can be a minority stock holder but be untouchable as Chariman. Then you can waste huge amounts of resources on unprofitable newspapers to satisfy your personal desire to be a player and engage in shameless nepotism. Just ask Rupert (if he doesn’t cut short the questions at the AGM).
Perhaps trying to create an atmosphere where life long learning is encouraged?
Pundits who call for labor reforms that would make it easier to sack bludging workers…
Richard, I know this is how labour market deregulation gets portrayed in the media and by the trade union movement, but do you really think that is what it is all about? Perhaps calls for labour market deregulation is a cry from employers to be able to actually MANAGE their workforce, so it is less costly to reorganise their labour resources so as to be able to deploy new technologies and/or work processes which can raise labour productivity? After all, particularly if there are a lot of sunk costs in switching to new techniques of production, this can be a risky thing to do.
In that sense, I agree with you that individual workers are constrained in the extent to how much they can actually become more productive. Bonuses and ‘performance based’ remuneration for employeees are only likely to have a minimal impact, if any. However, aren’t employers also constrained by labour market regulation in terms of the costs it adds to reallocating/shedding labour resources? Too often, trade unions expect job numbers, wages and conditions to be effectively preserved in aspic, preventing managers from effectively managing and entreprenuers from innovating.
Interested in your thoughts.
Also, what do you think about Franklin Fisher’s objections to the existence of aggregate production functions (and therefore, by implication aggregate productivity measures)?
I recall your cold shower being predicted, just not with such an unappealing label.
“It also seems to be ethically abhorent (almost literally creating second class citizens)” Almost, but not quite, because guest workers aren’t even citizens.
“Given I still favour a system that allows employees to move around and I dislike employer monopsony, I can’t really think of a way to foster this “on the job” human capital more than is being done so.”
Well, isn’t it nice when it turns out there is something we don’t have to worry about reforming. But surely an improvement would come with more wage and conditions flexibility.
“How do we get firms to improve productivity as it become available.”
that’s the wrong question. The question for that particular “we” is: How do we make sure that stupid barriers are not put in the way of improvements in productivity?
I’d be in favour of higher fees for tertiary courses that purport to train people for specific jobs! If the majority of university graduates had good generalised learning/research/analytic/self-discplinary skills and hence ready to take on a range of career possibilities, firms would have little choice but to accept that they would have more responsibility for on-the-job training. And if it’s really true that firms are wary of this because they’re concerned about the investments they might be making won’t pay off if workers just go elsewhere then is there a role for tax subsidies? E.g. perhaps employee incomes could be 100% tax deductible for 6 months if the employees have only just completed tertiary qualifications?
Wiz, salaries are already 100% tax deductible for your employer, even Uni grads, even those who studied lesbian whale lit.
Well then, 160% deductible :-)
I suppose one problem with my suggestion is what should count as ‘tertiary qualifications’, as obviously there’s a percentage of the population for whom a university or even TAFE education is probably not the best use of resources.
You don’t need to be overly concerned with the structure of the education or training- people will usually respond positively to a chance for improving their skills.
Even assisting numeracy and literacy skills pay off and feeds into apositive workplace culture.
Which will result in Maccas offering a 50% premium on standard wages for arts graduates.
Yes but the key word there is ‘usually’. You’re right though, we probably need more targetted solutions rather than broad policy suggestions.
Governance – how to get good governance? Yes a very tricky business since one can’t do anything much directly. Competition does the trick, but how do you bring that about. I think you try to get information flowing as well as possible, and we’ve barely tried that. When Ken put up his list of productivity enhancing policies, I wanted to add my “Windows on Workforces” to the mix. In any event I think it would improve management, perhaps quite a bit. And it would improve competition for good workplaces – and there’s a fair bit of evidence that workforces that people rate highly are more productive.
And I agree with a fair bit of what Djoof says.
“I’ll also disregard my feeling that ultimately productivity growth is beyond individual government policies and wealthy countries like Australia are sticking close to a frontier that is determined by the global stock of human knowledge…”
Thats an odd-ball thought bubble. The restoration of sound policy would lead to gains in productivity immediately.
“It is far too tempting to reduce productivity to labor productivity, and then to reduce that to the effort and ability of workers…”
You MUST reduce productivity to labor productivity or you will make a hash of things. What I see going on here is the curse of the one best metric. Where economists think that they have to sort out a single you-beaut formula, rather than using several, so they try and stuff a single metric chock-full of labor productivity and capital productivity and they only confuse themselves.
“I’ll also disregard my feeling that ultimately productivity growth is beyond individual government policies and wealthy countries like Australia are sticking close to a frontier that is determined by the global stock of human knowledge…”
Thats an odd-ball thought bubble. The restoration (or at least the advent) of sound policy would lead to gains in productivity immediately.
“It is far too tempting to reduce productivity to labor productivity, and then to reduce that to the effort and ability of workers…”
You MUST reduce productivity to labor productivity or you will make a hash of things. What I see going on here is the curse of the one best metric. Where economists think that they have to sort out a single you-beaut formula, rather than using several, so they try and stuff a single metric chock-full of labor productivity, capital productivity, and all sorts of other gear, and they only confuse themselves. Productivity of capital naturally falls with capital accumulation but rises with improvements in technology and other factors. If you have an abundance of capital goods quite naturally they spend most of their time lying around. There is no great tragedy with this. Technological enhancement naturally occurs with capital goods update and so is therefore embedded in the progressive actions of the economy if policy is sound.
More in keeping with the real purpose of the topic I concur that there are serious agency problems with our public companies. Just the other day I heard Alan Joyce emphasizing the discipline that Qantas executives had shown in terms of not grabbing for themselves pay rises and large bonuses. If it requires discipline for management NOT to stick its hands in the till, then we are some ways from supply and demand, merit-based earnings, natural law, and the harmonization of all interests in the globally benevolent market. I would argue that the top three layers of management of pretty much all our larger public companies have their hands in the till. And the point is well made that these resources might otherwise have gone into capital investment.
How do we solve the problem?
Well the problem of agency is partly to do with corporate law, partly a factor of human nature, and partly the result of non-market influences that favor larger over smaller firms. Regulation compliance costs being but one example. To solve matters we want to rejig our tax and regulations to bring down the average size of companies, and to enforce a default position where bigshot salaries will quickly deteriorate unless the shareholders vote forcefully (perhaps with a two-thirds majority) to maintain or enhance them. Other than that agency problems are a good argument for progressive taxation but only in the context of near-minimalist government.
The other measure I forgot to mention is to make limited liability mean 100% equity finance. This would improve resource allocation via the improvement of share pricing, bring down the average size of firms, harmonize the interests of society and the shareholders, and make the management far more beholden to the shareholders. In all these ways the problem of agency would be severely reduced.
Yes, the difficulty and complexity of running a company does bloat the size of companies and restrict entry. So simplifying things could help a little, though simplifying things . . . well that’s not so easy.
No-one’s been able to do it!
It simplifies matters greatly if limited liability means 100% equity finance, there is no company tax, and share buyback is allowed. That way the management job is pretty clear. Maximize the ratio of book value/outstanding shares. Always run a profit. If the shares are undervalued use some of that profit to buy-back shares.
Currently companies almost never try to maximize that ratio in the context of falling company size and revenue. This is a misallocation of resources just the same as paying bigshots too much money (in lieu of capital spending) is. And note also that the failure to have this as the main financial goal is also yet another factor making our firms on average too large. We have too many barely profitable companies, and loss-making companies, trying to get larger or stay the same size. Whereas they ought to ruthlessly be making a profit; even if that means growing smaller.
The company tax sets up irreconcilable conflict of interest and confusion in the goals of management and has to be eliminated post-haste. But that does not mean that we cannot have a 1% total assets tax with a threshold and a 1% total revenues tax with a threshold. Such taxes, if the threshold is large enough, do not have the same ghastly effects as a tax on profit does.
The agency problem, now I under why the high share price companies are the big ones?
Yes NG, solutions are easy to see, hard to grasp. The only big businesses I’ve worked in a big law firms, but whenever I see productivity annoyances it usually stems from the management organising things to make their job easier. Funnily enough, they tend to not be interested in helpful suggestions for change.
Management can get away with a greater level of agency when there is absolutely no way to gauge their performance or to say what the hell they ought to be doing. The company tax (among other things) makes it impossible to say what they ought to be doing and how well they are doing it. No-one knowing how management ought to be serving the shareholders leads to management serving themselves instead. As you would expect.
“The company tax (among other things) makes it impossible to say what they ought to be doing and how well they are doing it.”
So, an obligation to pay tax on profits and gains makes it impossible to gauge whether management are serving the interests of shareholders? Okey dokey.