CLUELESS ECONOMISTS?
Gautam Bhattacharya
Professor of Economics
KIMEP University
Almaty, Kazakhstan
And
Associate Professor of Economics (Retired)
University of Kansas,
Lawrence, Kansas, USA
October 26, 2017
I am indebted to Dr. Nadeem Naqvi for some lively discussion which led to the first draft of our joint paper and for graciously allowing me to use the material from our joint paper. I am also indebted to Dilobar Kassymova for providing research assistance.
In my classes, I sometimes show off different categories of economists (boasting that economists are not all dull and/or socially awkward!). I show them handsome economists (yes, they do exist!) athletic economists, superrich economists, transgender economists and many others. The most fun I have is when I show them a “clueless economist”!
Let me explain !
In mid 1990’s, a young researcher from Princeton wanted to test whether competitive markets in real life exhibit a uniform price (Graddy (1995)). She picked New York’s Fulton Fish market – a large wholesale fish market – selling mainly to restaurateurs and grocery stores, where a large number of independent stores operated from the same premises. Collecting a lot of data on sale prices and quantities and running several statistical tests, she found that the prices overall are the same for the same quality of fish. A little annoying item was that she found one ethnic group, Asians (and Koreans in particular), who were systematically discriminated favorably, meaning they usually paid a lower price than others!
It was a competent piece of research. However, the FBI took a keen interest in her research , they were already investigating Fulton Fish Market for collusion and other organized crime related activities! A little later the FBI accused the sellers of collusion, and soon afterwards the entire fish market mysteriously burned down, with all their records totally charred! (NYtimes (1995))
The reality, as I see it now, was that all the wholesalers were tightly controlled by organized crime, prices were not competitive but reflected almost perfect collusion, and the unfortunate Asians (Koreans) that bought cheaper fish were owners of inner city grocery stores in NYC area – they were paying protection money to the aforementioned organized crime group – so they got a little break on the fish prices!
I am sure the reality became obvious after the market burned down in 1996. Indeed, she mentions the involvement of organized crime in her later writings. And of course, I would speculate that she was not totally unaware of this at the time the data were collected and the paper was written (circa. 1991-93). But only the Federal Bureau of Investigation had inside information on this, not an academic economist.
Indeed, even in a market economy, sometimes things appear to be very different from what they really are.
When applying game theory to analyze equilibria in a more complex economy, the possibility of appearing to be clueless is a lot more serious. In a subgame perfect equilibrium, credible threats are usually not executed, so one might construct an incorrect hypothesis where credible threats are absent altogether. In political economy, where complex social and economic interactions are studied, one usually finds policy makers and academics who are blissfully naïve (and clueless)
My colleague, Nadeem Naqvi and myself have been perplexed by the “Stans” – we have been working in one of them for the last few years. On the surface, these “Stans” and other off-shoots of the Soviet republic are doing fine. Some have more resources than others. The BANK and the FUND periodically publish research monographs about foreign investment, business environment, GDP growth and other key economic indicators – assuming, erroneously, that these are budding capitalist economies. There is a steady stream of econometric work trying to predict efficiency of different monetary policy instruments and fiscal stimuli to promote economic growth
But the reality is, these are all “network” states, the “network” being a combination of government officials and others in power. Fiscal and monetary policies have superficial, at best marginal effectiveness in these countries.
The network state collects tributes legally by taxation and illegally by other means from business enterprises. They wield the threat of confiscation or loss of services. In return they provide order which is a combination of infrastructure and protective services. These are the important instruments, not traditional economic policy!
However, the “Stans” are not collapsing en masse, like the Soviet Union did. Indeed, some “Stans” are doing better than others. Some exhibit moderately high rates of growth and considerable private investment and business activity, whereas others only have the basic retail and other non-traded goods provided by the private sectors ! Is it because of their resource base? Or is it an outcome of a more complex game? Clueless we might be, but since we know some game theory, we tried to build a theory of “Stans” to show why some of them are more vigorous than others.
A little background here:
In the 70-year lifetime of the Soviet Union (1922-91), (i) there were virtually no property rights by which individuals had to live, largely because there was extremely limited private ownership of property (beyond one’s own autonomous labor), (ii) other than sale of labor time (hours or days for which labor power was sold), all other sources of income for a household were effectively prohibited, (iii) there were no businesses that households could buy and sell, nor of course any stock markets in shares of private firms, and (iv) there was no developed real estate market.
This was radically different from the economic base (of the relations of production and the associated inter-household distribution of income) on which the capitalist economic systems of the world were organized at the time. Moreover, resting on this economic base in the Soviet Union, the entailed legal and political superstructures pertaining to individual rights over property were virtually nonexistent. This is a huge contrast with the successful capitalist countries, in which the legal structure dealing with property rights was firmly in place, and there were political institutions that collectively were in continuing support of such rights under the law.
By 1991, there was rebirth of fifteen post-Soviet countries, and in these the corresponding collective consciousness of the people was nothing like that in the successful capitalist countries.
What kind of progress have the Group of 15 made towards the state’s provision of greater protection of property rights of individuals, especially to the businesses they give birth to and nurture into successful and profitable enterprises, simply because the greater such protection, the more fertile (profitable) will be the domain of the economy over which private business investment will be made in larger magnitudes, and in higher-return ventures? The larger the annual private business investment, the faster is the rate of growth of the economy’s capital stock, to which corresponds a faster rate of growth of real GDP. To avoid the fate of the late Soviet Union, these countries must adopt a path of economic growth supported by private investment. The data available in these countries are not entirely reliable, the data on private sector capital formation is not available for some, in others it fluctuates wildly. Further, sometimes the network state embarks on a “prestige” project and private investment occurs as an ancillary to this project. Once the project is completed, private investment goes down substantially.
| A snapshot of the Group of fifteen, 2016 | ||||||||||||||||||
| GDP Per Capita | Gross fixed capital formation, private sector (% of GDP) | |||||||||||||||||
| GFCF as % of GDP | ||||||||||||||||||
| Country Name | GDP per capita 2016 ($) | 2012 | 2013 | |||||||||||||||
| Lithuania | 29972 | Not available | Not available | |||||||||||||||
| Estonia | 29313 | Not available | Not available | |||||||||||||||
| Russian Federation | 26490 | 17.6 | 18.5 | |||||||||||||||
| Latvia | 25410 | Not available | Not available | |||||||||||||||
| Kazakhstan | 25145 | 25.7 (in 2006) | not available | |||||||||||||||
| Belarus | 18000 | 30.1 | 32.6 | |||||||||||||||
| Turkmenistan | 17485 | 10.5 | not available | |||||||||||||||
| Azerbaijan | 17439 | 11.8 (in 2007) | Not available | |||||||||||||||
| Georgia | 10044 | 17.5 | 16.9 | |||||||||||||||
| Armenia | 8621 | 22.2 | 19.4 | |||||||||||||||
| Ukraine | 8305 | 17.2 (in 2009) | not available | |||||||||||||||
| Uzbekistan | 6563 | 17.5 | 18.0 | |||||||||||||||
| Moldova | 5328 | 22.6 (in 2006) | Not available | |||||||||||||||
| Kyrgyz Republic | 3521 | 24.1 | 22.9 | |||||||||||||||
| Tajikistan | 3008 | 4.0 | 5.5 | |||||||||||||||
| Data on GDP and GFCF from various UN publications |
It is clear that in terms of GDP per capita, some of these countries have performed a lot better than others. The data on Gross private capital formation is not satisfactory, but again we can see a large difference in these numbers between the group of fifteen. Several other countries in East Europe, Asia and Africa, may also satisfy our criterion of network states.
Gross capital formation as a percentage of GDP varies wildly, ranging from 4% in Tajikistan to about 30% in Belaraus and 25% in Kazakhstan. (for comparison , GFCF is about 15% in USA which has a very large capital base, but it is about 40% in China and 25% in India) Why is it that some network states are more successful in promoting private investment than others. Is it only because they have more natural resources. Or is the reality more complex?
We applied a little bit of agency theory to look at a game between the “network” state and the agents who may be “qualified” or unqualified”
2. Model
The government (network) provides legal order S and collects revenue (tribute) T. Order includes infrastructure for enterprises plus a threat level for confiscation. This is essential for collection of tribute.
To sustain high return enterprises, the government must provide an order level sH while the level of order needed for a low-return enterprise is sL. There are two types of agents:
- “Qualified” – who either possess human capital or resources or have enough “network” connections to operate high return enterprises. They earn more from high return enterprises
2. “Unqualified” who earn more from operating low-return enterprises.
An unqualified agent may not be interested in operating a high-return enterprise, whereas a qualified agent has a choice of operating either a high-return or a low-return enterprise.
The objective of my exercise is to see under what conditions the equilibrium outcome will involve a situation where both high and low return enterprises are operated by the agents in the economy. This equilibrium is called a “productive” equilibrium. The other equilibrium is called a “stagnant” equilibrium because only low-return enterprises exist here, operated by both qualified and unqualified agents.
The payoff (surplus) from operating a high return enterprise by a qualified individual is
V = max [U(θ, S ) – T]
Where θ is the agent type (θQ qualified, and θU unqualified), sL is the order provided to low-return enterprises and sH is the order provided to high-return enterprises, and T is the tribute that must be paid to the government (tL to operate a low-return business, and tH to operate a high-return business).
Thus, if a qualified agent participates in a high return enterprise, his return is
VQH = U(θQ , sH) – tH
If he participates in a low return enterprise, his return is
VQL = U(θQ, sL) – tL,
Therefore he will choose to invest in H if VQH > VQL , and in L otherwise.
An unqualified agent, I assume, has a very low (possibly negative) return from investing in a high quality enterprise, so his return from investing in L is
VUL= U(θU, sL) – tL,
He will invest if VUL is non negative.
The main point of distinction between qualified and unqualified agents is
U(θQ, sH) – U(θQ, sL) > U(θU, sH) – U(θU, sL),
which implies that the qualified agents have a higher differential payoff between high return and low return enterprises.
The government maximizes a convex combination of the cost of providing order and the revenue earned,
G = – a[C(sL) + C (sH)] + (1- a)(tL + tH)
where a is the weight placed on the costs. Here G is the government’s objective function.
By assumption, it is more expensive to provide sH because it needs better infrastructure and possibly higher level of coercion so that greater revenue could be generated from the high return enterprises.
2. Equilibria
We look at three possible equilibria (scenarios). The proofs are in my joint paper with Professor Naqvi (Bhattacharya and Naqvi (2017))
1. The government provides only sL, enough to sustain low-return enterprise only. It is not profitable for the government here to provide sH, the revenue generated is the entire surplus of the agents (or the minimum surplus if the unqualified agents are diverse). This is the stagnant equilibrium. Here, the government does not find it optimal to provide sH because qualified agents will continue to choose low-return enterprises because for them the additional net return from operating higher return enterprises is negative.
2. The government provides both sL and sH, and collects different revenues tL and tH from different agents. Here, qualified agents choose high-return enterprises and unqualified agents choose low-return enterprises. The revenue earned tL is equal to the surplus of the unqualified agents – thus the unqualified agents are left with what we call only “normal profits”. The revenue earned tH from the high-return sector is less than the surplus of the qualified agents, because they self-select to operate high-return enterprises. So the qualified agents earn more than “normal profits”. This is the productive equilibrium.
3. The third scenario occurs where the government provides only sH and extracts all the surplus by collecting tH. Here, the low-return enterprises are not functional. We call this a “Highly-productive” equilibrium. This type of equilibrium will not exist if there is a large number of unqualified agents in the economy, or if it is not expensive for the government to provide order level sH. Thus, this is the kind of equilibrium that a society should strive for under improved governance and availability of opportunities.
For the kind of economies we are considering here, this equilibrium is unlikely to exist because there is likely to be a large number of unqualified agents who will find operating a low-return enterprise profitable even if their surplus is close to zero.
3. Remarks
1. What exactly happens in “stagnant” equilibrium?
This equilibrium is achieved as an outcome of voluntary interaction between a government which provides order and infrastructure and agents who decide to operate enterprises that they find profitable. In this equilibrium, although there are many qualified agents, they self-select to operate low return enterprises because the return from doing so is higher than operating high return enterprises. The government provides only order level sL because provision of sH is redundant in this case.
2. What exactly happens in “productive equilibrium”?
When is the productive equilibrium likely to exist? In a productive equilibrium, qualified agents operate high-return enterprises and unqualified agents operate low-return enterprises. Naturally, the level of private investment will be higher in this equilibrium.
I have proved that this equilibrium will exist when the proportion of qualified agents is neither too small nor too large. As we mentioned before, it is unlikely to be too large, so “not too small” is the relevant criterion. This proportion depends on the level of human and natural resources at the disposal of the agents and size and strength of the “network” in the economy
The second result I prove is that the productive equilibrium will exist when going from low to high order level increases surplus proportionally more for qualified than for unqualified agents.
4. More Remarks
We observe, in Central Asia, Eastern Europe, in south Asia and in Africa, different kinds or regimes where appropriation is the norm – both regimes with relatively vigorous economic activity and regimes with relative stagnation. This formulation might shed some light on why some regimes appear to be more stifling than others.
Did I just present you with a well-disguised tautology? Possibly!
So let me check again – what exactly is the implication of all this?
Basically, even in a network society, agents do not curl up and surrender all surplus to the network. Nor does the network brutally destroy the society. The outcome depends on the proportion of agents who are educated, connected to the network, and willing to take the risk of eventual confiscation of appropriation of their entire investment. If enough of these agents exist, and the network finds it not too expensive to provide the right infrastructure for high return enterprises, then the economy will exhibit a productive equilibrium where substantial private investment may occur. On the other hand, if this proportion is small, the network state will only build an infrastructure to support low-return enterprises. In a country like Tajikistan, for instance, private investment and economic activity are on a much subdued level than the relatively successful “Stans”. Each of the countries in this group has a varying amount of natural resources that they can export. But for sustainable growth, they all need development of vigorous economic activity. My model indicates that the proportion of qualified agents, and the “network’s” willingness to provide a higher level of infrastructure and property rights are some of the key elements for sustainable growth.
REFERENCES
- Bhattacharya, G. and N. Naqvi, (2017), “A Theory of Appropriation under Self-selection by Agents” Mimeo. KIMEP University, Almaty, Kazakhstan.
- Graddy, Katherine (1995), “Testing for Imperfect Competition at the Fulton Fish Market”, The RAND Journal of Economics, Vol. 26, No. 1 (Spring, 1995), pp. 75-92
3.Fire Sweeps Through Major Building at Fulton Fish Market
By JOE SEXTON
Published: March 30, 1995