A rapid increase in interenterprise arrears (IEA) is one phenomenon in transition economies that has generated lively debates among policy-makers as well as economists. Most studies have approached the IA issue from two directions. The first, a “financial discipline” approach, argues that IEA stems mainly from the lack of financial discipline, and hence can be reduced by enforcing financial discipline on enterprises. 2/ The second approach, a “credit crunch” approach, focuses on underdeveloped credit markets and tight credit policies in transition economies as major underlying causes of IEA accumulation. A group of Russian economists advanced a simpler version of this approach by proposing credit injection as the solution (Aleksashenko, 1992). A more sophisticated version argues that credit injection, essentially a bailout, would complicate the arrears problem by weakening the financial discipline of enterprises, and instead, seeks a more permanent solution through massive privatization and the introduction of effective bankruptcy regulations, in line with the financial discipline approach (Calvo and Coricelli, 1993 ).
To provide further insights into the IEA issue, this paper develops a general equilibrium model of interenterprise arrears, characterized by n-stage production technology with random productivity shocks. The general equilibrium model incorporates interactions of suppliers, buyer firms and consumers and analyzes their reactions to productivity shocks and policy variables. The model suggests that (i) the existence of IEA is consistent with optimal behavior of firms and consumers; (ii) substantial business risks in transition economies contribute to high levels of IEA in those countries; and (iii) IEA could increase as a result of rapid privatization and commercialization at least at the early stages.
Our model offers two distinct advantages. First, our general equilibrium model, by incorporating interactions of market participants, allows one to analyze the end result of policy effects rather than only initial reactions. For example, while higher real interest rates on bank credit would encourage buyers to accumulate trade payables and delay payments, the increased opportunity cost of trade credits would discourage suppliers from extending trade credits. As a result, the net effect on the equilibrium level of trade credit will depend on the price elasticity of demand. Another example can be seen from the regulatory imposition of prepayments for goods delivered and services rendered. The model suggests that it will be difficult for the government to enforce prepayments since a seller, say, a rubber producer, may not insist on prepayments from a buyer, say, an automobile company for fear of losing its clients.
Second, our model, by introducing of n-stage production technology, allows us to study the effects of privatization and commercialization on IEA. Contrary to conventional views that privatization will help reduce arrears by enforcing financial discipline, the model shows that privatization and commercialization of state enterprises could increase IEA by delineating the financial responsibilities of formerly vertically integrated production units and thereby increasing the volume of market transactions per output. The model implies that highly industrialized countries like Russia are likely to accumulate more arrears as a share of GDP than largely agricultural transition economies such as Albania.
Section II presents the general equilibrium model of interenterprise arrears. Section III derives equilibrium trade credit and arrears from the model. Section IV examines the effects of industrial structure and privatization on IEA. Section V discusses the relevance of the model to IEA experience in the Russian Federation in 1993 and 1994. Section VI discusses the model’s policy implications.
II. A Simple General Equilibrium Model With N-Stage Production Technology
This section presents a general equilibrium model of interenterprise arrears, characterized by n-stage production technology with random productivity shocks. Firms, while taking prices and the risk of productivity shocks as given, decide how much input to buy, output to sell, and trade credit to provide to buyers in order to maximize profits. Consumers, living two periods, decide how much to consume and save and how much time to spend on work and education to maximize their utility over their lifetime. Competitive equilibrium prices are determined by technology, cash-in-advance constraints, and government policy variables as well as the money supply, while an equilibrium real interest rate is determined by efficiencies in the accumulation of human capital.
The production technology in the economy is characterized by n stages of production, with constant returns to scale. The final product at time t (yt, n) is produced indirectly from the primary input, xt, through n production stages in the following way: The first intermediate good yt, 1 is produced from the primary input xt, owned, accumulated and supplied by consumers. Then, the second intermediate good yt, 2 is produced from yt, 1 and so on, until the final product yt, n is produced from the (n-1)-th intermediate good, yt, n-1.
Hence the production function for the first intermediate good is yt, 1 = B1xt, where B1 is the marginal product of the primary input x and the production function for the i-th intermediate good is yt, i = Biyt, i-1, where i > 1. Similarly, the production function for the final product can be expressed as yt, n = Bnyt, n-1, where Bn is the marginal product of an intermediate input yt, n-1.
We assume that the productivity variables, Bi’s, are subject to random shocks as follows:
For simplicity, we assume that all pi’s are equal to p, a business risk parameter of the economy.
Then the law of large numbers ensures that with an infinite number of firms in each industry, the actual final output yt, n is equal to the expected output, which is a function of a primary input xt, technology parameters Aj and the risk parameter p, as follows:
Also, actual output yt, n in terms of actual output yt, i may be expressed as follows:
Firms in each industry purchase inputs, produce goods and sell them at time t in accordance with their production relations to each other. Industry i represents a production stage i described above and thus firms in industry i purchase inputs from the (i-1)-th industry, produce yt, i and sell the product to firms in industry i+1. Firms are risk-neutral and are assumed to under perfect competition.
Firms are engaged in financial transactions during their normal production process. Buyer firms borrow cash from banks at a market interest rate, rt+1 and utilize the borrowed money to pay for inputs at time t. Bank loans must be fully paid at time t+1. Seller firms deposit the sales receipts at banks at time t and earn interest income at a rate of rt+1 payable at time t+1.
Instead of borrowing cash, buyer firms at time t may use trade credit from suppliers to purchase intermediate goods. At time t+1, they repay the trade credit in cash to suppliers at a default rate p, which is the probability of random shocks known to both buyers and sellers. The default rate, p, could be interpreted in two different ways: firms are equally subject to the same productivity shocks p; or p percent of firms will surely fail to pay but they are not identified prior to transactions. When defaulted, debtor firms are charged penalties by the disciplinary authorities (government) and penalties are confiscated from the capital of debtor firms as government revenues, some of which will be reimbursed to seller firms.
At time t, each firm intends to maximize profit expected at time t+1 taking prices and interest rates as given. 3/ Before a productivity shock occurs, a representative firm in industry i chooses an optimal amount of intermediate inputs (yt, i-1) supply (si) and demand (di) for trade credit and money demand (mi) which maximize the expected profit. Then the objective of a representative firm in an industry i is to maximize:
where si is trade credit receivable (trade credit granted by the firm) per unit of sale; θi, the government reimbursement rate per foregone receivables; di and mi, trade credit payable (trade credit owed to suppliers) and demand for money per unit of purchase; and γi, a penalty charge per overdue payment of firm i. By definition, θi and γi are between zero and one. For notational convenience, we define yt,0 = xt and pt,0 = Pt, x
The first term on the right side of the equation represents expected revenues at time t+1. Revenues are expected from three sources: sale revenues net of default loss, interest income from cash deposited with banks, and government compensation for arrears in trade receivables. The second term is the total expected cost of production, composed of three elements: payments for intermediate goods, interest payments for bank loans at interest rate rt+1, and penalty charges by the government for arrears in trade payables.
In profit maximization, firms face two types of financial constraints. First, firms in an industry i must pay cash or use supplier credit as expressed in the following condition:
Second, firms need to make immediate cash payments for at least a certain portion of purchases. The cash-in-advance constraint for firms in an industry i can be expressed as following:
The cash-in-advance constraint is determined by industry-specific elements such as production cycles and durability of products and by economy-wide institutional elements such as social norms, financial development stages, laws and economic systems. For example, services such as transportation generally require instant payment for the whole value of the services rendered, whereas consumer durable goods do not require such payment. Also, many products can be purchased on credit in developed countries but not in undeveloped countries. In addition, the cash-in-advance constraint can be essentially zero in planned economies since planning authorities always provide non-cash payment instruments (credit lines) to enterprises to settle transactions authorized by the planner. We assume that no arrears are allowed in wages, that is, β1 = 1.
The solutions are straightforward, given the linear nature of the optimization exercise. When the cost of borrowing money (rt+1) is greater than the cost of receiving trade credit (p(γi-1)), the expected profit is an increasing function of trade payables and a decreasing function of money. 3/ Therefore a firm, as a buyer, demands a minimum amount of money (mi and a maximum amount of trade payables (di), that is, m*i = βi and d*i = 1-βi. At the same time, the firm, also as a seller, wishes to decrease trade receivables si to its minimum since the opportunity cost of trade receivables (rt+1) is greater than the benefits (p(θi -1)). However, if the firm provides trade credit to its clients less than the equilibrium level, its clients will switch into other sellers. Hence the firm’s optimal supply of trade credit s i* is determined at the level which equalizes profit margins of all firms to a minimum sustainable level, which is zero. The optimal level of trade credit s i* satisfies the following first-order condition for maximizing profit:
The economy consists of an infinite number of identical consumers, who live two periods. 4/ There is no population growth. At time t, a representative consumer born at time t maximizes the following two-period utility function:
where ct, t and c t, t+1 represent consumption during period t and t+1 of the consumer born at time t and, δ is a subjective utility discount rate.
Consumers face two different types of constraints. First, we assume that consumers may not take consumer credit and thus face the following cash-in advance constraint:
Second, consumers face budget constraints. All consumers born at time t are endowed with human capital affecting productivity of labor. Also consumers can use one-unit of time in each period of their life. Human capital (ht) accumulates in proportion to time spent on education in the following way:
where B represents efficiency in human capital accumulation. Consumers have different budget constraints depending on their ages. When young, they can devote ν units of time on education and (1-v) units of the time for earning wages. The only source of income is wages. Thus, the budget constraint for the younger generation can be expressed as:
where bt+1 is demand for time deposits. When old, consumers spend all of the time working and consume the total income. The sources of income are wages and financial income from deposits saved in their youth:
where rt+1 is a nominal interest rate on deposits between time t and t+1.
Given the utility function and the two constraints, a representative consumer chooses consumption, financial asset holdings, labor supply and educational investment to maximize utility, taking prices as given. The first-order condition for maximizing utility takes the following form:
which implies that nominal interest rate of the economy is:
where πt+1 is an inflation rate which can be measured by a growth rate of nominal wages (pt+1, x/pt, x. This equation implies that the real rate of interest is determined only by the technology variable B. For notational simplicity, we define rt+1 = r.
From the first order condition (12) and budget constraints (10) and (11), the optimal time spent on education, v*, is:
which shows that time spent on work and supply of effective labor depend on the technology and preference parameters.
The government’s only revenue comes from the penalty charges on debtor firms for trade payables in arrears. The government maintains a balanced budget by reimbursing a certain portion of trade receivables in arrears to creditor firms and transferring all remaining revenues to the firms in a lump sum. Also, the government may change the money stock by distributing new money to firms free of charge.
5. Market-clearing conditions
In this economy, a competitive equilibrium is determined as follows. The firm’ maximization problem yields demand functions for intermediate inputs, labor, money and trade credits, and supply functions for output and trade credit. Likewise, the consumer’s maximization problem yields demand functions for consumption goods and money, and a supply function for the primary input. The equilibrium prices are then obtained from the market clearing conditions.
First, in an equilibrium, aggregate money demand from enterprises and consumers should be the same as money supply, MSt, as follows:
where, as noted earlier, βn+1 (= β0) is one, that is, no consumer credit is allowed.
Second, market-clearing conditions for trade credit are:
which means that aggregate demand of the i-th industry for trade credit from the (i-1)-th industry is equal to its aggregate supply by (i-1)-th industry to i-th industry.
Third, the financial market is in equilibrium at time t if the following condition is met:
The left side of the equation represents supplies of financial assets (time deposits) by young consumers born at this time and firms, and the right side represents demands for financial assets from old consumers born last period and firms.
Fourth, under the assumption of no population growth, market-clearing conditions for primary input and final consumption goods are as follows:
Finally, under the assumption of a balanced budget for the government, the resource constraint of the closed economy can be derived from the market-clearing conditions (or the sum of equations for expected profits over all firms) as follows:
Under the previous assumptions that no consumer credit is allowed(β0 =1) and that no delays in wage payment are allowed (β1 =1), the above equation can be rewritten as:
which means that the sum of total income of consumers in the economy is equal to their total expenditure on final goods (or the total value of the national product). Then, a combination of equation (1) and (21) yields the price of the final product relative to that of the primary input as:
This equation implies that the relative price depends on the technology parameters Ai and the random shock parameter p. Increases in productivity will raise the relative price of the primary input, since one unit of the primary input can produce more units of the final product. On the other hand, an increase of p will decrease the relative price of the primary input, because of more losses during production.
Then, equation (23) can be modified to express the price of the 1-th good relative to the price of the final product as follows:
Note that according to equation (2), relative output (yt, n/yt, i). Also, Ji(r) represents the relative value (pt, i, yt, i/pt, nyt, n) because from equation (2) showing quantity relations and equation (24) showing price relations, it follows that the value of good i can be expressed in terms of the value of good n as:
Finally, when equation (1) is plugged into equation (25) and the modified equation is subsequently plugged into the money market clearing condition (equation (15)), the equilibrium prices of the products can be fully expressed in terms of exogenous variables as:
Accordingly, the equilibrium prices are functions of technology, a productivity shock, cash-in advance constraints, an interest rate, and government policy parameters as well as the money supply.
III. Equilibrium Trade Credits and Arrears
1. Aggregate trade credit and aggregate gross arrears
The equilibrium level of aggregate trade credit, Tt, is calculated as a sum of equilibrium trade payables of each firm as follows:
Consider a special case where cash-in-advance constraints are the same across industries (βi’s are equal to β for i=1...n) and the government reimbursement rate for forgone receivables in an industry (θi-1) is the same as the penalty charge per overdue payments in the same industry (γi). In this case, the following holds:
Then the aggregate trade credit can be simplified as:
Since buyer firms can not repay trade credit when they fail to produce output, the equilibrium level of gross IEA of the i-th industry and aggregate gross IEA will be trade credit multiplied by the probability of productivity shocks:
The solutions for equilibrium trade credit and gross IEA have the following implications. First, a certain amount of real gross IEA exists unless all βi’s are equal to one, and the equilibrium level of aggregate gross arrears is determined by a risk parameter for productivity shocks (p), a cash-in-advance constraint (β) and an industrial structure (n).
Second, the government’s compensation (θ) of creditor firms or charges of penalty (γ) on debtor firms cannot wholly eliminate all the gross arrears. The administrative compensations and charges have limited effects on equilibrium trade credit and mostly pass through to changes in equilibrium prices of products. In particular, when n is large enough, the policy variables do not affect aggregate trade credits and consequently gross arrears. In this case, aggregate gross arrears can be approximated by:
Third, changes in money supply do not affect equilibrium gross arrears as a percentage of the money stock.
2. Net arrears
Net arrears of the i-th industry are calculated as a difference between trade payables in arrears and trade receivables in arrears as follows:
When Ki(r) is plugged into equation (33), the net arrears can be expressed as:
The equation for equilibrium net arrears shows that as long as the i-th and the (i+1) - th industry have different cash-in-advance constraints, net arrears of firms in the i-th industry also do not have to be zero in equilibrium. This suggests that non-zero net arrears reflect a difference in the cash-in-advance constraint across industries rather than a persistent existence of inefficient or inviable firms.
Equilibrium net arrears will be small if the difference in β is small. In the special case where all the β are identical, net arrears can be zero. However, even in this case, gross arrears remain positive since profit-maximizing enterprises have both trade receivables and trade payables.
IV. Commercialization, Privatization and IEA
One of key characteristics of industrial structure in the former socialist economies is that firms usually contain more vertically integrated units than in market economies (Nove, 1980 and Berliner, 1976). In an extreme case where a whole economy is one firm composed of a number of vertically integrated units, there is no need for payments among the production units and thus no demand for trade credits in the economy. When some of the vertically integrated units become autonomous firms, that is, when commercialization takes place, then the autonomous firms will begin to ask other autonomous firms to pay for supplied goods and services, and an unpaid portion will be treated as trade credit. However, commercialization alone will not necessarily result in the accumulation of trade credit. Commercialization will increase trade credit in the economy only when transactions among firms are no longer automatically financed by the government, that is, only when commercialization is accompanied by privatization.
The relationship between trade credit and privatization can be highlighted in the following formula for equilibrium trade credit in the simple case shown in the previous section. Along with commercialization and privatization, the equilibrium level of aggregate trade credit will increase from zero to:
The above formula shows that an equilibrium trade credit depends on the number of production stages (n) which is usually associated with the industrial structure and the development stage of an economy. It implies that countries with more complicated industrial relations like Russia could accumulate a higher level of IEA relative to GDP than reforming countries with a simple industrial structure such as Albania. Also, the formula shows that privatization increases trade credits. Chart 1, derived from equation (35) under an assumption of a constant growth of the number of privatized firms, illustrates that the level of trade credit increases as privatization proceeds. The more rapidly the privatization proceeds, the greater the level of trade credit will be at an initial stage of privatization. Eventually, the level of trade credit will be solely determined by cash-in-advance constraints (β).
Chart 1.Trade Credit and Speed of Privatization
A constant speed of privatization is assumed and privatzatization is represented by an increase of the number of firms.
Beta refers to a share of cash payments in total value of purchases.
V. Lessons from Russia, Based on 1993 and 1994 Experiences
1. Persistent arrears
The Russian economy first encountered the interenterprise arrears problem in 1992, when prices were substantially freed. Gross arrears grew rapidly in Russia: from less than 4 percent of GDP in January 1992 to approximately 25-40 percent of GDP by July 1992 (Ickes and Ryterman, 1993). After IEA were reduced at the second half of 1992 through mutual setting, IEA resumed the increase in early 1993 in nominal terms and rebounded in the fourth quarter of 1993 in real terms measured by the CPI (Table 1). 5/
|Industry||(End of period; billion rubles)|
|Deflated by CPI)||3,882||3,230||3,129||2,764||2,304||3,256||3,507||3,793||3,012|
|Overdue Trade Receivable||1,798||2,422||4,184||6,956||12,342||21,558||30,588||41,632||48,921|
|(Deflated by CPI)||1,798||1,286||1,320||1,157||1,311||1,632||1,883||2,171||1,716|
|(Deflated by CPI)||2,738||2,358||2,276||2,344||2,402||2,745||3,104||3,247||2,568|
|Overdue Trade Payable||1,139||1,672||2,623||5,209||9,162||16,367||24,872||34,806||41,451|
|(Deflated by CPI)||1,139||887||828||866||973||1,239||1,531||1,815||1,454|
|(Percentage change over previous period)|
|Overdue Trade Receivables||…||35||73||66||77||75||42||36||18|
|Trade Receivables Deflated by CPI||…||-17||-3||-12||5||12||8||8||-21|
|Overdue Trade Receivables Deflated by CPI…||-29||3||-12||13||24||15||15||-21|
|(End of period; ratio)|
|Trade Receivables/Ruble M2||0.64||0.68||0.63||0.78||0.86||1.11||0.96||0.94||0.89|
|Overdue Trade Receivables/Ruble M2||0.30||0.27||0.27||0.33||0.39||0.56||0.51||0.54||0.51|
|Trade Receivables/Annualized Output||0.11||0.10||0.11||0.12||0.13||0.18||0.21||0.20||0.15|
|Overdue Trade Receivables/Trade Receivable||0.46||0.40||0.42||0.42||0.45||0.50||0.54||0.57||0.57|
|Industry, Agriculture Construction and Transportation|
|Trade Receivables (billion rubles)||…||…||…||22,634||35,957||57,986||79,294||102,899||122,951|
|Overdue Trade Receivables (billion rubles)||…||…||…||8,904||15,918||27,729||39,790||57,381||69,233|
|Trade Receivables/Ruble M2||…||…||…||1.07||1.13||1.49||1.33||1.33||1.28|
|Overdue Trade Receivables/Ruble M2||…||…||…||0.42||0.50||0.71||0.67||0.74||0.72|
|Ruble M2 (billion rubles)||6,051||8,913||15,767||21,171||31,799||38,823||59,621||77,353||96,378|
The Russian government reacted to the resumption of arrears accumulation In a variety of ways in 1993 and 1994. The government issued a number of normative acts aimed to monitor arrears positions, clarify payment procedures and strengthen payment disciplines. For example, “File No. 2” was reintroduced to the banking system in October 1993 to measure the arrears positions of firms after its initial abandonment in June 1992. 6/ In addition, payment priorities were established for various obligations, and firms were obliged to sell foreign currency deposits to settle overdue payments, according to a presidential decree of May 1994. 7/ Another presidential decree of the same month prohibited each enterprise from keeping more than one ruble settlement account with resident banks as a way to facilitate monitoring of firms’ compliances to the May decree. 8/ Also, the Federal Bankruptcy Commission under the State Committee for the Management of State Property (GKI) set up specific bankruptcy criteria in May 1994 including the size of overdue debts relative to liquid assets. 9/ The commission has identified thousands of potentially bankrupt firms, and enterprise managers began to face an increasing threat of losing jobs. Managers were to be personally responsible for violations of regulations on payments and accounting disciplines, and a manager of Norsi, an oil refinery company, was fired in the summer of 1994 as the first prominent case. 10/ Furthermore, the central bank and commercial banks improved payments system substantially to a level largely comparable to the international standard although some inter-regional settlements continued to lag. As a result, no enterprises could accuse the payments system of responsibility for overdue payments.
The policies were generally ineffective in 1994 in reducing arrears. Trade arrears and trade credit grew much faster in 1994 in real terms deflated by the CPI than in 1993 (Table 1). Trade arrears of industrial, agricultural, transportation and construction companies reached around 6-8 percent of annualized monthly GDP in the second half of 1994 from about 4-5 percent in the same period of 1993, and trade receivables to 11-14 percent from 10-11 (Table 2). 11/
Trade receivables and payables refer to those of industrial, agricultural, transportation and construction companies.
Trade receivables and payables refer to those of industrial, agricultural, transportation and construction companies.
2. Trade credit outgrowing output
Trade credit as a percentage of output has rapidly increased in Russia over the last three years. Trade receivables of Russian industrial enterprises grew from less than 2 percent of yearly sales in 1990 to about 20 percent in summer months of 1994 (Chart 2). Two questions arise: (1) what may explain the trends? and (2) is the level something to be concerned about?
Chart 2.Russian Federation: Trade Receivables and Output
Sources: Costomatst and IMF stiff estimates.
Our model provides one explanation for the trends. Commercialization and privatization took place in Russia at an unprecedented pace. 12/ As noted in Section IV, new autonomous firms created by the process began to ask other autonomous firms to pay for transactions which have been purely internal bookkeeping operations before, and thereby amplified aggregate demands for trade credit. Considering the high degree of vertical integration within Russian state enterprises, our model suggests that the resultant increases in market transactions must have been substantial, and also the growth of trade credit. 13/
An international comparison of trade credit shows that most, if not all, of the increase in trade credit in Russia may reflect its convergence to adequate levels of trade credit in market economies. Chart 3 shows that the 1993 level of trade credit in Russia was in line with those in market economies. Trade receivables of manufacturing firms are typically 10 to 25 percent of yearly sales (or 1.2 to 3 months credit) in five sample countries (Japan, the U.S., West Germany, Taiwan and Korea) and the ratios are more or less stable. 14/
Chart 3.International Comparison: Trade Receivables and Annual Sales
Sources: “Business Statistics” and “Flow of Funds” for U.S.; Goskomstat for Russia; and Bank of Korea for others. Data are based on nonfinancial nonfarm firms for U.S.; industrial firms for Russia; and manufacturing firms for others.
3. Price pass-through
There was an argument that the nature of trade credit in Russia is fundamentally different from those in market economies because debtor firms in Russia are not charged with high penalty interest rates while those in market economies are (Ickes and Ryterman 1994). However, our model strongly suggests that implicit interest charges may pass through to prices. Although it is hard to find conclusive evidence of the price pass-through in Russia at this stage, there is some evidence suggesting that prices normally charged by enterprises might include a component reflecting expected losses from arrears as predicted in our general equilibrium model.
For example, a deputy minister of atomic energy has recently announced that the sector’s enterprises are ready to make a 30 percent reduction in electricity tariffs for those who guarantee full payments. 15/ Similar statements were made by a director of Rosugol (the largest state coal company) and the Ministry of Railroads. 16/ Recent efforts of major upstream industries (coal, electricity, fuel and metallurgy) to create a price cartel may reflect the prevalence of the price pass-through practices of enterprises facing increasing opportunity costs of overdue trade credit. 17/ Under the cartel agreement, member enterprises are to charge lower prices among cartel members as one way to circumvent nonpayment problems among them. 18/
A report of the Fund provides a concrete example of the price pass-through. 19/ A fur processing factory was setting its prices by applying a margin to costs of production, taking into account the prices charged by other enterprises in the same sector and adding taxes. Markup rates were running at 50-70 percent of the full cost of production in early 1994, compared to pre-transition margins in the range of 20 percent.
4. Net arrears positions of enterprises
The Russian experience suggests that net arrears are relatively small compared to gross arrears. A mutual netting process in 1992 revealed that the sum of net IEA of net debtor enterprises was about one-sixth of the sum of total gross IEA on July 1, 1992. 20/ More recent information on net arrears in overall economy is not available but arrears statistics in sectoral levels suggest that such a pattern of arrears positions remains. For example, construction enterprises could not pay Rub 1 trillion on time to suppliers as of end-1993 since contractors did not pay Rub 4.2 trillion for their construction work. 21/ The pattern indicates that most enterprises with overdue payables also have overdue receivables and cautions against the presumption that net IEA positions of a firm be interpreted as a definitive criterion of the firm’s solvency. 22/
A broad policy implication of the model is that a certain portion of interenterprise arrears is likely to persist for a while in transition economies. Trade credit as a share of GDP will increase as vertically integrated enterprises are broken up during privatization and commercialization. The size of overdue and unpaid trade credit will be affected by, among other things, overall business risks which are generally high in transition economies and cash-in-advance constraints which tend to converge to those in market economies. Thus, penalty charges and other measures designed to enhance financial discipline will reduce aggregate arrears but will not eliminate them completely. At the same time, monetary expansion through either direct lending to debtor firms or government guarantees of bills of exchanges will lead to inflation without any substantial reduction in real arrears.
It should be emphasized that our model makes conventional assumptions about firms’ behavior and thereby excludes two types of firms which could also contribute to arrears accumulation and would obviously reduce welfare of a society: firms with principal-agency problems and bankrupt firms. For example, a director in a ceratin firm might be willing to provide trade credit to other firms without any repayment capabilities in return for a personal kick-back when such behavior is not likely to be detected and penalized. Also, some obviously bankrupt but unclosed firms may continue to receive goods in credits when full government bailouts are widely expected by suppliers. Our model shows that arrears could persist even when the two obviously malign cases of arrears are excluded.
Assumptions of the model can be modified without affecting the major implications of the model. First, the current model assumes that the default rates are determined by exogenous productivity shocks only. However, debtor firms certainly have strong moral hazard to default if they think that they can get away with using exogenous productivity shocks as an excuse not repaying trade credit. Such intentional default will be prevalent when business information is deficient or very costly to obtain, or when penalties for such default are insignificant. Arrears of such origins could be permanently reduced by policy measures to oblige firms to keep key business information accurately and transparently. Accounting and auditing reforms would be helpful in that respect. It would also be helpful to develop financial intermediaries which are more aware of credit history and financial status of firms than other institutions. However, both measures could take some time to be fully implemented. Establishing a workable collateral system could be one of the most practical and effective steps to reduce IEA.
Second, an assumption of price flexibility can be replaced by an alternative assumption of rigid prices. In that case, effects of monetary policies on macroeconomic variables will be different from those suggested in the current model. For example, in the current model, a sudden monetary contraction affects prices only. However, when the money stock declines but prices do not, demand for materials will decline while inventories increase and as a result, output may decline ultimately. Although our model could be modified in such a direction, we do not explore the case further since we think that such a disequilibrium case, while plausible in the short term, is unlikely to be sustained.
AleksashenkoS. “Tight Financial Politics in Russia: the First Experiment and its First Results” (translation from Russian), Boprosi Ekonomikil (December1992).
BerlinerJoseph S.The Innovation Decision in Soviet Industry (1976).
BigmanDavid and SergioPereira Leite “Enterprise Arrears in Russia: Causes and Policy Options” (Washington) International Monetary FundWorking Paper No. WP/93/61 (August1993).
BrechlingF.P.R. and R.G.Lipsey “Trade Credit and Monetary Policy” Economic Journal (December1963).
CalvoA. Guillermo and FabrizioCoricelli “Credit Market Imperfections and Output Response in PCPR” (mimeo) (1993) Vol.1.
CalvoA. Guillermo and FabrizioCoricelli “Output Collapse in Eastern Europe; the Role of Credit” (Washington) International Monetary FundStaff Papers Vol. 40 (March1993).
CalvoA. and Jacob A.Frenkel “From Centrally Planned Economy to Market Economy” (Washington) International Monetary FundStaff Papers (June1991).
Central Bank of RussiaTekushchie Tendintsii v Denzhno-Kreditnov Sfere (Current Tendencies in the Monetary-Credit Sphere), No.1 (1994).
CliftonV. Eric and Mohsin S.Khan “Interenterprise Arrears in Romania” (Washington) International Monetary FundStaff Papers Vol. 40 (September1993).
FerrisJ. Stephen “A Transaction Theory of Trade Credit Use” Quarterly Journal of Economics (May1981).
IckesBarry W. and RandiRyterman “Roadblock to Economic Reform: Interenterprise Debt and the Transition to Markets” (mimeo) (1993).
IckesBarry W. and RandiRyterman “The Interenterprise Arrears Crisis in Russia” Post-Soviet Affairs (1992).
LafferArthur B. “Trade Credit and the Money Market” Journal of Political Economy (March1970).
NadiriM.I. “The Determinants of Trade Credit in the U.S. Total Manufacturing Sector” Econometrica (XXXVII) (July1974).
SastryA. S. R. “The Effects of Credit on Transactions Demand for Cash” Journal of Finance (September1970).
Board of Governors of the Federal Reserve SystemFlow of Funds Accounts (Various Issues).
Goskomstat of the Russian FederationFinance in the Russian Federation (1991).
Goskomstat of the Russian FederationNational Economy of the Russian Federation in 1992 (1992).
Goskomstat of the Russian FederationStatistical Bulletin; Selected Economic Indicators of the Russian Federation for 1992 (1993).
Goskomstat of the Russian FederationSocial-Economic Situation of Russia (Various Issues).
We are grateful to Messrs. John Anderson, Daniel Citrin, Donal Donovan, Richard Hass, Ernesto Hernández-Cáta, Keon Hyok Lee, Michael Marrese, Donald Mathieson, and Jeronimo Zettelmeyer for helpful discussions and comments on an earlier version of this paper.
See, for example, Bigman and Leite (1993), who propose reasons for IEA accumulation and discuss schemes to reduce IEA.
The model does not exclude the possibility that the cost of borrowing money (rt+1) is less than the cost of receiving trade credit (p(γi-1)) and in that case, the solutions will be opposite. However, we would not consider this case sustainable since in such case all banks will be out of business. Alternatively, we may assume that productivity of the economy is high enough to guarantee positive real interest rates. See equation (13) for determinants of real interest rates.
This is not a restrictive assumption. The introduction of the infinitely-living agents does not change the main result of this paper.
See IMF Economic Reviews, Russian Federation, 1993, pp.30-31, for detailed information on the results of the mutual setting.
Presidential decree No. 1662 of October 19, 1993 “On improvements in settlements in the national economy and their timely executions to increase accountability.”
Presidential decree No. 1005 of May 23, 1994 “On additional measures to normalize settlements and strengthen the payment discipline in the national economy.” In a move to implement the decree, the Central Bank of Russia (CBR) established an interim procedure how legal entities might use funds in their settlements and current accounts (CBR letter of June 30, 1994, No. 98). Also, the State Tax Inspection, the Ministry of Finance, the CBR and the Ministry of Justice issued a resolution to support the decree (Government Resolution No.682 of September 8, 1994).
Presidential Decree No. 1006 of May 23, 1994.
Government Decree No. 498 of May 20, 1994.
Kommersant Weekly, October 18, 1994, p.22.
A CBR publication suggests that arrear statistics prior to June 1992 are not directly comparable to those statistics afterwards; until June 1992, when balances in settlement accounts of debtor enterprises were not large enough to cover arrived payment orders, the differences were automatically transferred to the File No.2 whereas afterwards, the arrears criteria were basically set up by contracts between creditors and debtors (Tekushchie Tendintsii v Denzhno-Kreditnoy Sfere (Current Tendencies in the Monetary-Credit Sphere), No 1, CBR, January 1993).
For example, the number of large firms that decided to commercialize increased from 429 to 1,952 in the first quarter of 1993 alone. During the same period, the number of large firms that completed commercialization increased from 80 to 367 (source: GKI).
On average, 48% of products produced by Russian heavy industrial enterprises are products of enterprises in other industrial sectors (source: authors’ estimates based on GKI data).
Source: Business Statistics and Flow of Funds for U.S; Finance in the Russian Federation for Russia; Financial Statements Analysis for other countries, A sharp decline of the Korean ratio in 1972 reflects major regulatory changes on curb markets in that year.
Rabochaya Tribune, September 10, 1994.
Rabochaya Tribuna, September 16, 1994.
Both the CBR refinance rate and the interbank reference rate were negative in real terms during most of 1993, but then turned sharply positive toward the end of year and into 1994.
Government Resolution No.1024, September 9, 1994.
Russian Federation, Report on National Account Statistics Mission, June 1994.
The sum of the net arrears was Rub 402 billion whereas the sum of total arrears position of enterprises was Rub 2,300 billion as of July 1, 1992 (Economic Reviews. Russian Federation, 1993).
Ekonomika i Zhizn, p.11, No.21, May 1994.
A recent government decree (No.498, May 20, 1994) stipulates arrears position of enterprises as one of bankruptcy criteria.