Saturday, February 18, 2017

What if Governments Owned the Robots?

What if wealth creation depended even more on robots than is already the case? More specifically, some have asked in recent years: who "should own" the robots? Yet something about shared ownership of robots misses the mark, in terms of monetary flows and in some instances, a possible lack thereof. Hence one important question in this context is, to what outcome does the automation process contribute, in the first place? Is that outcome a direct representation of wealth, or a further derivative of wealth? If the answer is the latter, robots and their associated automation may not contribute to output in ways that are (also) available for redistribution.

Some of the "dark matter" of these system outcomes could affect investment patterns as well. Timothy Taylor noted in a recent post that "80% of gross investment now goes to replace old capital, rather than add to the capital stock." What else is at stake besides the reduced costs of automation? To what extent is dark matter caught up in human investment? Meanwhile, the human capital deemed most "important", is sometimes more costly than the capital investment long associated with firm activity.

A post from Tyler Cowen prompted me to consider the robots ownership discussion. He asked, what is government in a world where robots do the work? After all, government (at least beyond the military hardware involved) is mostly the paid labour of humans who generate time based product. One MR commenter also asked: why not an option where everyone owns robots? Perhaps the relevant equivalent here, however, is simply ownership in stock of companies where automation does lead to verifiable gains in output and wealth creation.

In spite of any further elements one might attribute to robots, basically they're little more than automation processes which follow the traditional productivity path of the Solow Residual. Thus far, this has been the path used by institutions in both primary and secondary market activity. Nevertheless, robots can contribute to tradable sector output and definable wealth, while the same is not necessarily true for non tradable sector output in a secondary market position.

The dependent position of the latter, also means automation accrues to reduced budget costs that are not necessarily associated with wealth creation. Unfortunately, the automation that reduces costs, does not have to be associated with either increased or definable output. And the difference matters, if society attempts to utilize robots as a potential source of redistribution. Indeed, much of the capacity which made it possible to do so, may already have been tapped via the use of robots and automation which contributed to tradable sector wealth in the twentieth century.

Wednesday, February 15, 2017

Thoughts on Knowledge and Equilibrium Dynamics

Some aspects of economic equilibrium are relatively straightforward, such as measurable output and the ability of central bankers to provide reasonable monetary representation such as NGDPLT. However, the dynamics of time based service output versus tradable sector output, are mysteries which - at least for this blogger - are begging to be solved. In particular: how do today's revenue dependent sectors affect nominal income potential? If all forms of product are solely represented by money (i.e. time value lacks an economic platform so as to exist in a sustainable relation to itself) the Solow Residual could ultimately mean widespread deflation, for total factor productivity.

In other words: regardless of perceived value of labour or lack thereof; money as the sole representation of economic value, means more labour will eventually be excluded - wherever politically possible - in the foreseeable future. Dietz Vollrath provided a simple explanation of this equilibrium problem, in a post from a year or so earlier. He noted a technology driven decision making process, in which present day output is being maintained at a relative constant. Whereas the input for said output is gradually being reduced.

But why? For centuries, the productivity enhancing effects of the Solow Residual, guaranteed the importance of both inputs and outputs. It was reasonable to expect reliable growth gains, alongside increased dispersal and application of knowledge. Presently, however, the incredible efficiency of tradable sector activity, finally means far less labor force participation is necessary - even should tradable sector output be increased. Consequently, with less tradable sector income available for redistribution, there's growing pressure to reduce employment in sectors which operate by an altogether different set of equilibrium dynamics.

Meanwhile, aggregate output is being maintained via increasingly unorthodox monetary policy means, as lower labor force participation translates into a smaller customer base for a more direct support of total output. And when less input is required for total output, wages decrease in relation to total output as well.

However, in the past, less employment in relation to total output, meant employee pay gains. Why not now? When manufacturing activity was a larger GDP component, tradable sector activity provided wage gains through vastly increased output of final product. Whereas the time based final product of non tradable sector activity - since it cannot multiply the actual product of an individual's time - lacks the additional output which would normally translate into wage gains. Here, should productivity gains occur, they may be relegated to instances in which time based product can be diverted, to replication of knowledge on the indirect terms of automation or technology.

When non tradable sector time based product is asymmetrically compensated, there's a reduction in total factor productivity since other resource options are consequently reduced. Even though asymmetrically compensated time based product provides flexibility for large systems service coordination, the process involves considerable trade offs. In what is necessarily a permissions only context, time based product cannot create sustainable patterns via internal means. Paradoxically, since this form of time based product remains dependent on broad equilibrium conditions, the only way to gain gain greater "efficiency" is by reducing the amount of available time, for a product whose primary appeal may be directly linked to the time involved.

Given the experiential reality of human nature, further reductions of time based product don't always provide optimal solutions. For instance: when and how do higher income levels seek to minimize their own consumption of time based product? When do they prefer the time of others to automation? The answers are important for the consumption potential of all income levels, given the need for greater time based productivity. Should it turn out that people of all incomes appreciate the value of time based product, perhaps this fact needs some formal economic recognition, before the slow but steady removal of labour in the marketplace becomes even more pronounced.

More to the point: time based services are also needed, because of the knowledge they disperse, and the societal networks they both coordinate and maintain. Consider a local example, regarding growing backlogs of stored evidence for previous unsolved homicide cases. In some instances, tales of partially destroyed backlogs reach the news media. Is this due to the growing futility of gaining the time based resources those backlogs suggest? Even though many observers find the destruction of criminal evidence outrageous and unbelievable, this act is only one of countless others, re the economic reversal which was expressed by John Cochrane in a recent post:
How can economies forget? How is it that once we learn to do something better, that knowledge can be lost and economies move backward? How can productivity decline? Viewing productivity as knowledge, it would seem almost impossible for it to do so - and real business cycle theory was often derided on that point. Yet middle ages Europeans lost the recipe for concrete, and time after time we have seen economies get worse. How can our own productivity be growing so slowly overall when so much we see around us is progressing so fast?
Ultimately, productivity also needs to be measured in time based units, so that the productivity of knowledge linked time can finally evolve. Until this occurs, the demands of time based product on other resource capacity could mean more political peril.

Knowledge use will always be valuable, via the asymmetric compensation that societies can sometimes provide for their best and brightest. Still, this process alone is scarce enough to disperse, protect and promote the use of knowledge in society. Manufacture is actually becoming too efficient, to reliable serve as additional revenue for asymmetric compensation. Before additional output and growth can be regained, time value also needs to serve as a direct unit or point of wealth creation.

We need more than just the highest skilled and lowest skilled individuals in the marketplace of the future. The time of the average skilled individual needs its own productivity calculation and application, for the growth and preservation of valuable knowledge to endure. So long as the Solow Residual is the only means by which to capture productivity gains (and money the only means to capture economic value), the likelihood of less input for a stable but lackluster aggregate output, is likely to continue - along with an implied loss of nominal income.

Sunday, February 12, 2017

Apples, Haircuts, Growth and Output

What might apples and haircuts - as an imaginary economy with only one service and one product - suggest about equilibrium growth and output potential? A recent post from Nick Rowe, encouraged me to explore the dynamics and make some related connections as well. Here's Rowe's opening paragraph:
An economy produces two goods: apples and haircuts. The production function for apples shifts up or down every year at random, depending on the weather. The production function for haircuts never shifts. The weather causes relative prices to change. When there is good weather, and the apple harvest is large, the price of haircuts in terms of apples rises (the price of apples in terms of haircuts falls). When there is bad weather, and the apple harvest is small, the price of haircuts in terms of apples falls (the price of apples in terms of haircuts rises).
Importantly, weather defines the setting, as it establishes an output range for the initial product. Good weather translates into a good crop in which not only does the abundance of apples cause the price of each apple to fall, the greater abundance of apples overall allows the costs of haircuts to rise. Likewise, bad weather is a smaller crop in which each apple costs more, and there is less apple output in aggregate for the resulting haircut valuations.

The beauty of the one product one service economy, is that it illustrates what I've referred to as primary and secondary market functions of output and growth potential. Apples can be likened to the random resource capacity of primary markets, which provide settings for initial market conditions. Haircuts are the secondary market capacity which represents time based services in general. That said, however, the fact that time based product is dependent on primary markets is obscured at the aggregate level, by the still growing effects of Baumol's cost disease. Consequently, the lower productivity of time based services in its dependent position, instead of creating wealth via its own means, is gradually displacing the resource capacity of primary market functions. In some recent discussions re cost disease, a commenter at Scott Sumner's blog left this useful reference point from William Baumol's book on the subject:
In the long run, wages for all workers throughout a country's economy tend to go up and down together. Otherwise, an activity whose wage rate falls behind will tend to lose its labor force. Auto workers and police officers will see their wage rise at roughly the same rate in the long run, but if productivity on the assembly line advances while productivity in the patrol car does not, then the cost of police protection will rise relative to manufacturing. Over several decades the two sectors differing cost growth will add up, making personal services enormously more expensive than manufactured goods.
If today's time based service products are dependent on primary market formation - in ways that are mostly being papered over via extensive debt, taxation and hidden unemployment - what mechanism created such a massive problem in the first place?

This circumstance can actually be traced to money as representative of all resources and marketplace output - be those resources scarce and finite such as time or land product, or the sum total of all product capacity. Yet one only needs to recall that the value of random resource aggregates continue indefinitely to pull away, from the scarce and finite product of time and land. Nominal income is embedded in the latter, in ways that provide a mirror for the valuation of random resource capacity. Various marketplace participants stake income claims ahead of other would be participants in the economy, simply because they can. Even though the subordinate position of time value has price elasticity which responds normally to output in the above one product one service economy, the sticky wages of a complex equilibrium are reinforced by land values which already mirror local aggregate income levels.

Since time value does not (yet!) have formal representation as a store of value, it doesn't have a specific, economically quantitative template whereby production gains can be generated. Instead, today's time value has little "choice" but to exist in relation to total marketplace output. Even though the prices of haircuts (as a secondary market, i.e. not internally generated) rise and fall with apple output in a simple economy, it's difficult for anyone to determine a "reasonable" time based product value, in relation to the full resource potential of an otherwise productive and complex economy. To add to the confusion, a vast majority of income is ultimately stored in real estate - a process which particularly makes time value resistant to good deflation and production norms.

Again, return to the thought of money as the sole representation of marketplace output as a beginning point. The reason that time value aligns with total resource output, is the fact that it does not (presently) have the option of aligning production capacity in relation to itself. Not only would internal facing of services generation overcome Baumol's disease, this organizational capacity would move into a primary market position that generates new wealth. Over time, time based services as a primary market position, would increase total growth and output. Once tradable sector activity experiences the massive relief of higher labor force participation and a strong customer base, one could even hope that today's political chaos will subside.

There's another problem for money as the sole representation of total output value. Of late, some seek to downplay the role of money in macroeconomic activity - in part because it does suggest hard equilibrium limits beyond the ways in which today's "maker" roles are conceptualized. Indeed, secondary market formation is becoming more self limiting, as the Baumol effect creates more distortion for budgets and debt formation.

Nevertheless, I believe that time value as expressed on formal, primary market economic terms, would be a better approach to long term growth than using debt to sustain an excessive amount of secondary market patterns. Some people think about these things and despair. So it confuses others, when I'm still able to get excited about the potential for a better economic future. Is it still possible to convince anyone that low labour force participation and reduced growth potential can be addressed via a marketplace for time value? I certainly hope so.

Thursday, February 9, 2017

Skills Use Permissions as a Malthusian Constraint

Economies that rely extensively on skills permissions will eventually run into problems, if those permissions include basic time based functions - especially during times of economic stagnation. Permissions based time value can easily become a Malthusian constraint, in part because this organizational structure leads to a fixed product which is relatively inelastic. Is it possible for individuals of widely varying income levels to live well, without time based product that requires such a high level of skills permissions? Answers may depend on whether other structural options exist, that can taking the finite nature of time value into consideration.

Even though skills permissions have evolved for good reasons (people want well qualified health providers for instance), constraints may become Malthusian, if organizational capacity for time based product remains mostly permissions oriented. Nevertheless, it's difficult to think about supply constraints in context, when time based product becomes an outsized component of marketplace costs. Those costs also imply a broader marketplace range, than what may actually exist.

The time based product of healthcare, is likely more of a Malthusian constraint than is recognized. Recently, Dietrich Vollrath explained how Malthusian constraints might hold:
First, living standards are negatively related to the size of population. This would occur if we had some sort of fixed factor of production. Typically, one might say it was agricultural land, but you could just say resources if you like. It isn't even important that they are truly fixed. So long as the resources are inelastic, whether due to a physical limit or because bringing them into use is prohibitively expensive, you'd satisfy the first characteristic of a Malthusian economy.
Malthus was "proven wrong" so to speak re his original argument, when advances in technology made it possible for agriculture to overcome the tremendous food production constraints he witnessed during his lifetime. However, even though healthcare has also greatly benefited from technology, the underlying problem is the way healthcare is positioned as a secondary market structure. Even though technology could make some aspects of time based healthcare production less expensive, that fact alone won't dramatically affect the revenue sources by which healthcare providers gain incentive to increase output. The harder it is to access revenue gains in general equilibrium, the more difficult it is to pass the lower costs of technological progress, to consumers.

On the other hand: not only did technology greatly increase agricultural output; the reality of vastly more food, benefited other crucial marketplace factors at the same time. That - in turn - meant revenue growth in aggregate from multiple sources. Technological expansion of agriculture was a major first mover - or primary market - event. In order for healthcare practitioners to accomplish anything remotely similar, the process would include providing assistance for institutions which structure aggregate time value as a central feature for replication and output gains.

In other words, time value would function as a basic commodity for further knowledge and skills processing, in a first mover market position. Think local services "refineries" perhaps, only with less costly production procedures! Instead of pipes and metal running everywhere across the landscape, human neural networks would make the crucial connections. Or perhaps one could refer to these processing plants as "farms" for new areas of productive agglomeration. Ultimately, the product of additional economic time value would provide benefits to other marketplace components simultaneously, much as more food meant earlier gains across entire economies.

While time based product will remain a fixed factor of future production, those limits need not exist to such a degree that services production and consumption face arbitrary limits. Today, time based product cannot easily replicate itself, without facing further dilution of the external revenue sources it requires. Fortunately, replication of services product is also possible via internal wealth creation, through cooperative time processing. By organizing for greater services output which need not rely on already existing revenue sources, time based product would be able to overcome some of the Malthusian constraints that presently exist.

Monday, February 6, 2017

Jobs Are Also a Cost. How to Respond?

"Also" is part of the title, as I recently emphasized investment as a cost. Given the fact today's organizations completely depend on the Solow Residual to realize production gains, both public and private institutions can be expected to return to this basic principle - in spite of the occasional diversionary tactic such as Trumponomics.

Consequently, how to generate more output via fewer (costly) jobs, even while building an environment which could provide means for individuals to remain economically viable? While I continue to stress a marketplace for time value as an option, a recent post from Mark Perry encouraged me to consider some specifics. Perry (of AEI) highlights several individuals in his arguments against "jobism", and I'll include a previous quote from Richard McKenzie:
Job creation (and protection) is a favored goal of our leaders because it appeals to existing political interests and is seductively misleading and counterproductive. It is also one of the easiest goals to achieve. To create or kill jobs, all Congress has to do is to obstruct progress and kill or retard opportunities for competitiveness and entrepreneurial spirit.
Ultimately, we need to juxtapose traditional job creation with mutual forms of reciprocity, on formal economic terms. Given the slowing pace of worldwide economic growth, the employment constraints of the Solow Residual especially compel us to look for new options. Mutual employment would be different from the jobs offered by present day institutions, in part because efforts to create social and economic value would extend both ways.

Mutual employment means more choice for all concerned, than offering one's time for hire to institutions that are in search of specific skill sets. One reason for the versatility of time arbitrage, is the fact one's time (for services production and consumption) is like one's own money to spend. The process is backed by newly generated money, if and when others accept one's offer via their own newly generated time commitment. Each unit of time becomes a new source of commodity wealth, which allows compensation to take place via internal means.

A marketplace for time value, would allow individuals to transform time value for services into a formal means of economic quantification. What's particularly important is the fact this process doesn't rely on money as the sole means. Should time value gain the ability to serve (alongside money) as a medium of exchange, unit of account and medium of account, new incentives for economic growth would be unleashed. Participants would gradually be able to build welfare enhancing nets of social support, which can be difficult to realize through monetary representation, alone.

In some respects, it's difficult for money to provide safety net functions for all incomes at a comprehensive level. Why so? Money represents the totality of what is both fixed and random resource capacity. This wide range of resource capacity continues to grow, in relation to the fixed scarcities of time use potential. If that weren't problematic enough, a wide range of time aggregate potential, is only being lightly mined by today's institutions.

Even though money - as the sole representation of all wealth - has provided tremendous value as an economic tool, it lacks the depth to capture the full potential of time use. Even though time value has infinite potential for product differentiation, its fixed scarcities make it difficult for time product to compete directly with other resource potential. Consequently, individuals find themselves compelled to select from time production and consumption options which feel oddly limited, in contrast with the possibilities our minds suggest.

Time value as a measure of wealth, would more closely mirror the variety and differentiation, which a massive degree of investment in human capital has made possible. The gains in product quality over time from a fixed point of time aggregates, would provide production gains in a context which isn't really at odds with the Solow Residual, as is today's uncertain services quantification. Granted, the time we choose to commit to others, would remain a cost in the ultimate equation. But it's a cost that we ourselves would choose to bear, from the resource capacity which would already be ours to spend.

Sunday, February 5, 2017

When are Divisions of Labour an Adaptive Strategy?

One reason this question is important, is the fact that, since Adam Smith's time, divisions of labour are central to the stories we've told ourselves about progress. For tradable sector product, divisions of labour have meant greater productivity over time, in the form of both increased output and product quality. Nevertheless, something has been missed in this recounting: labour divisions for non tradable time based product, don't translate into the same productivity gains.

Consequently, even though divisions of labour contribute greater quality to time based product, they have not generated output gains in time based product formation. This current institutional reality, in turn limits services complexity to the most productive regions of the world. Worse, some non tradable sector labour divisions are so investment intensive, that their knowledge use restrictions contribute to polarization and societal division.

Why has this occurred, and what might be done? First: product which is directly linked to time, is incapable of self replication. The output gains which are possible are those which exist elsewhere, by which other individuals contribute to the creation of more time based product in a similar capacity. However, as the reimbursement of time based product is currently structured (via secondary market formation), participants depend on the same pool of potential revenue, for time based service formation in a given equilibrium.

Since the present organizational capacity of time based product remains revenue dependent, its representative associations are often reluctant, to approve separately existing replication that results in expanded output. After all, if revenue cannot be internally generated (to expand output), expanded output mostly means diluted revenue to compensate those who already provide important knowledge and time based services. Perhaps due to the focus on product quality by only the "most qualified" providers, output loss has been treated as an inevitable result, by all concerned. Nevertheless: for the time based product of non tradable endeavour, hard divisions of labour have proven to be a poor strategy for the preservation of valuable knowledge and skill.

This form of non tradable sector activity inadvertently divides labour in ways which - instead of promoting societal progress - increase class divisions for knowledge use which only grow wider with the passage of time. Yet the intent of labour divisions by tradable sectors was never "set in stone". From the beginning, divisions of labour were intended as flexible tools by which institutions could adapt to changing consumer preferences and needs. Indeed: if management or workers had insisted on hard and specific definitions for knowledge or skill to produce tradable sector product, it's difficult to imagine how the Industrial Revolution would have ever taken place.

Divisions of labour are not an adaptive strategy for institutions to evolve and grow, when quality is demanded by extensive investment which in turn limits the quantity of time based product that is possible. Fortunately, the divisions of labour which contribute to time based product, could become more flexible in the future. But in order for this to occur, personal time value needs to more closely reflect the resource capacity which people actually have at their disposal. Symmetric compensation and coordination, sets the template for aggregate time value to operate as a true marketplace: one in which individuals can discover the time based product they are most comfortable with.

Best: the fact that wealth can be internally generated through symmetric compensation processes, is actually a side benefit. By placing the possibility of time negotiation on a common platform, individuals can use time value for evolutionary and adaptive gain, much as divisions of labour have been utilized in tradable sector activity for as long as anyone can recall.

When there is no marketplace in which time value can exist in relation to its own resource capacity, time aggregates must be compensated from other system components. As asymmetric compensation proceeds apace, individuals often find themselves compelled to rely on debt, as the sole means to honor time based commitments and responsibilities.

So long as economies grow rapidly, it's possible for individuals to adjust for vast differences in asymmetrically compensated time value, through additional debt loads. But when economic stagnation sets in, the "more output for less labour" Solow residual, acts like a slowly tightening noose, gradually undermining the possibility of full employment, social inclusion and continued growth. Even though the Solow Residual means production benefits for tradable sector activity, there is no "more output for less labor" equivalent for time based product. If there are not enough individuals participating in the output potential of time based product, ultimately there are too few consumers for either time based product or tradable sector product.

It's time to begin thinking differently about the present organizational capacity which insists on hard divisions of labour for time based product. Otherwise, non tradable sector activity will not be able to adapt and evolve in sustainable ways. The global economy is in desperate need of nations looking inward, to see what they have been doing wrong at home, instead of obsessing over the industries that have succeeded by putting flexible divisions of labour to good use for the right reasons. A marketplace for time value would not only protect domestic employment at home, but also the vast quantities of knowledge use and skill which are part of the present day global economy.

Saturday, February 4, 2017

The Limits of Indirect Reciprocity

Of indirect reciprocity, Shawn Parrish recently wrote:
One of the behaviors that humans display a lot is "indirect reciprocity". Distinguished from "direct reciprocity", in which I help you and you help me, indirect reciprocity confers no immediate benefit to the one helping. Either I help you, then you help someone else at a later time, or I help you and then someone else, some time in the future, helps me.
In recent centuries, indirect reciprocity has especially made it easier for populations to coordinate a wide range of activity, among individuals who would scarcely be able to coordinate such activity on direct terms. However, there's a problem. As the wealth divide between aggregate time value and the value of all other resource capacity continues to grow, there are corresponding difficulties in coordinating living and working arrangements, between a wide range of income levels. For instance, even though prosperous regions can remain open to tourists and visitors, it's less of a simple matter now, for them to make permanent space for those with low incomes. One important reason, is that individuals with smaller income are less able to contribute to the high costs of either extensive infrastructure building, and/or maintenance.

More than ever, populations need different levels of infrastructure design and commitment, so as to maintain both direct and indirect reciprocity with one another, on economic terms. Many of life's important challenges can be met by those with smaller income levels, so long as provisions remain in place for design capacity which allows more innovative and less costly infrastructure. Otherwise, much of the indirect reciprocity that populations once took for granted in prosperous regions, will only become more difficult to maintain over time. Unfortunately, when wide income variance leaves too little room for basic forms of social and physical infrastructural options, trust is one of the first casualties.

Still, it's possible to generate more direct forms of reciprocity, for the time based product which is now in short supply. Doing so, would help to address the growing disconnect between incomes at a structural level. Not only does today's disconnect increase societal distrust; this vast income divide makes it appear as though millions are incapable of contributing either to their own destinies, or the destinies of others. However, in order for people to gain greater means for reciprocity - whether direct or indirect - they need the freedom to do so. Today, it is this freedom, that is missing. And without the right to contribute to our societies the best we can, others will continue to think that we are not capable of doing so.