Resilience and Growth in the Small States of the Pacific
Chapter

Chapter 14. The Role of Excess Liquidity and Interest Rate Pass-Through for the Monetary Transmission Mechanism in the Pacific

Author(s):
Hoe Khor, Roger Kronenberg, and Patrizia Tumbarello
Published Date:
August 2016
Share
  • ShareShare
Show Summary Details
Author(s)
Jan Gottschalk

Monetary transmission in the Pacific is complicated by the absence of a number of channels that are available to central banks in advanced economies. This leaves monetary policy dependent on interest rate pass-through, which has been ineffective in the region, largely because of the nature of its financial markets rather than issues with the monetary policy frameworks. Very little can be done in the short term to improve the functioning of the interest rate channel. In the long term, however, the available policy options could considerably improve the functioning of the banking sector, improving both competition and market infrastructure.

This chapter revisits discussions during the Sixth Seminar on Monetary Transmission Channels, Liquidity Conditions, and Determinants of Inflation, a regional central banking seminar cohosted by the Bank of England’s Centre for Central Banking Studies and the Pacific Financial Technical Assistance Centre. The chapter focuses on issues of excess liquidity and limited interest rate pass-through in Pacific island economies.

It does not provide a complete summary, but highlights selected points that help to better understand the role of these factors for the monetary transmission mechanism in the Pacific island economies. The focus is on the aftermath of the global financial crisis when the transmission mechanism was perceived to be weak, with excess liquidity and limited interest rate pass-through seen as signs of this weakness. The chapter seeks to recapitulate the discussion in an informal manner, adding literature references where needed for depth, especially with respect to interest rate pass-through.

Excess Liquidity

Defining Excess Liquidity

Commercial bank reserves held with the central bank can be decomposed into required reserves and free reserves.1 Central banks typically require commercial banks to hold a portion of their deposit liabilities with them for monetary policy purposes. These required reserves are sometimes, but not always, remunerated. Free reserves are the remaining reserves held by commercial banks with the central bank that do not contribute to fulfilling reserve requirements. In other words, they constitute the difference between total and required reserves.

Free reserves can be decomposed further into voluntary reserve holdings to meet a bank’s liquidity requirements and excess reserves.2 Voluntary reserves are additional reserves that commercial banks hold for precautionary motives; the demand for voluntary reserves arises from a desire to self-insure and can vary depending on the uncertainty of payment flows. Reserves in general are a highly liquid and safe asset, and the lack of financial investment instruments in the Pacific island economies can imply that banks may be holding reserves voluntarily, simply due to the lack of other liquid and safe investment opportunities that are readily convertible into cash without undue loss. From a monetary policy viewpoint, an increase in voluntary reserve holdings is not problematic as such (even though it could signal a rise in uncertainty). But monetary policy tends to be concerned about the level of excess reserves; that is, reserves that are likely to be in excess of what banks are holding voluntarily. The rest of this section reviews the extent to which excess liquidity is present in selected Pacific countries and what it implies for the monetary policy stance and transmission mechanism.

Measuring Excess Liquidity

Liquidity has risen significantly in the Pacific islands since the global economic crisis. Table 14.1 shows significant increases in liquidity since 2008 in countries with data on liquidity indicators. In 2013 banks in Fiji, Papua New Guinea, the Solomon Islands, and Tonga were holding reserves significantly above the regulatory level, and most held high proportions of their assets in liquid form.

Table 14.1Liquidity Indicators for Selected Pacific Islands 2005–13(Percent)
200520062007200820092010201120122013
Liquid-Assets-to-Total-Deposits Ratio1
Fiji171522162225282827
Papua New Guinea767678707371757171
Solomon Islands483621223556607774
Tonga212324242834445054
Liquid-Assets-to-Short-Term Liabilities Ratio1
Fiji211925202832393839
Papua New Guinea26161815181920
Solomon Islands73878080
Tonga
Loan-to-Deposit Ratio
Fiji778275878285767775
Papua New Guinea383736464750464750
Solomon Islands495980977458494443
Tonga979596947975687176
Liquid-Assets-to-Total-Assets Ratio
Fiji151319141820242423
Papua New Guinea636465585857595755
Solomon Islands393017162743456159
Tonga161818182326343631
Liquid-Assets-to-Required-Reserves Ratio
Fiji365230375282350265297281282
Papua New Guinea750848829520646553628579480
Solomon Islands6454882782974727678248381,062
Tonga173223289285655763952962979
Free-Reserves-to-Required-Reserves Ratio
Fiji605417129130104132122120
Papua New Guinea1711255015311958825427
Solomon Islands1,9639842581178093,4434,0364,360554
Tonga9308443358442629771790
Sources: Central bank prudential data; and IMF, Pacific Financial Technical Assistance Center.

Liquid assets exclude nonmarketable securities.

Sources: Central bank prudential data; and IMF, Pacific Financial Technical Assistance Center.

Liquid assets exclude nonmarketable securities.

The extent to which there is clear evidence of excess liquidity differs across countries. As already noted, identifying excess liquidity requires attention to prudential as well as monetary policy considerations. The appropriate level of voluntary reserves depends on the level of risk, the predictability of deposit withdrawals, and the ease of liquidating nonreserve assets. As many recent papers note, the situation in Pacific financial markets suggests that, on most of these dimensions, banks should aim to hold voluntary reserves well in excess of required levels. The exact margin required, however, is hard to quantify as it depends in part on the perception of risk and the risk aversion of individual banks. With that in mind, the following provides an initial assessment of the presence of excess reserves for the four countries (Table 14.1):

  • In Papua New Guinea, loan-to-deposit ratios were low, indicating plenty of room to expand lending from a supervisory perspective. However, banks did not hold a large proportion of their nonlending assets in central bank reserves as they invested them in securities, largely central bank and Treasury bills, which are not marketable. Free reserves were about 30 percent above the level of required reserves, and overall liquid assets were about 20 percent of deposits and short-term liabilities. This was at the low end of what would be expected given market conditions in Papua New Guinea.
  • In Fiji, the loan-to-deposit ratio, the highest in the region on average during 2009–13, is at a comfortable level from a supervisory perspective, but significant increases in the ratio would require close prudential attention. Banks hold most of their nonlending assets as reserves and thus have adequate liquidity to manage short-term liabilities and shocks. But such liquidity does not appear excessive.
  • In 2013, the loan-to-deposit ratio in Tonga was higher than the ratio in Fiji, Papua New Guinea, and the Solomon Islands, and only slightly below that of Fiji on average during 2009–13. Banks held significant levels of free reserves with the National Reserve Bank of Tonga and had more than adequate liquidity to manage short-term liabilities. However, given the continued high level of nonperforming loans in Tonga, supervisors would expect banks to hold higher-than-normal ratios of liquid assets, and any sharp decreases in liquidity ratios or increase in lending volume should be monitored closely.
  • In the Solomon Islands, loan-to-deposit ratios are low, and banks hold large amounts of free reserves with the Central Bank of Solomon Islands. Liquidity is well above that needed to manage short-term liabilities, and so there is plenty of room to expand lending from a supervisory perspective.

Excess Liquidity and the Monetary Policy Stance

Interest rate issues arise when excess liquidity leads to an undesired easing of the monetary policy stance. This has been the situation in Papua New Guinea, where the interest rate on central bank bills has declined since around mid-2011, even though the Bank of Papua New Guinea preferred a tightening in its monetary policy stance, as indicated by subsequent increases in the Kina Facility Rate. This situation arises because of large foreign exchange inflows—in the case of Papua New Guinea, due to the commodity boom and foreign direct investment inflows to the mining and petroleum sectors—that lead to a substantial increase in liquidity unless they are fully sterilized. When the Bank of Papua New Guinea scaled back the volume of sterilization around mid-2011 (not least because of the sizable fiscal costs of these operations), this led to a sizable increase in commercial bank reserves with the central bank. The increase probably corresponded to a buildup of excess liquidity and a bidding down of rates on central bank bills as banks tried to invest at least part of their reserves in these instruments. The result was, in effect, an easing of the monetary policy stance.

In other countries in the Pacific, central banks desire a loose monetary policy stance, but perceive the presence of excess liquidity as a sign that monetary policy is ineffective because it does not lead to the desired increase in lending, which would absorb excess liquidity. In short, the conclusions are as follows:

  • The underlying concern that the monetary policy transmission mechanism is weak or ineffective is well founded. However, the main culprit is not excess liquidity in itself but a weak interest rate transmission channel that largely reflects structural characteristics of financial markets in the Pacific—for example, the absence of institutions such as credit bureaus that facilitate bank lending, and the small size of these markets.
  • The notion that excess liquidity would be eliminated if banks increased lending is true (and intuitive) for individual banks, but much less likely to be operative at the systemic level. The issue is that an increase in lending by one bank tends to create deposits—and therefore reserves—at another. This suggests that the level of excess reserves does not constitute a reliable summary statistic of whether monetary policy is effective. Lending volumes and interest rates are better indicators for this purpose.

The Role of Excess Liquidity in Monetary Policy Transmission

The clearest case for a strong role of liquidity emerges when banks are liquidity constrained. Injecting liquidity in such a situation will affect lending substantially by easing liquidity constraints. However, the institutional setup of monetary policy operations in advanced economies such as the United Kingdom or the United States ensures that such a situation practically never arises. That is, if banks are reluctant to lend in these countries it is not because they are liquidity constrained, but because the risk-adjusted return on lending is not high enough compared to the risk-adjusted return on other investments (including simply holding reserves). A different institutional setup in the Pacific could imply that banks may at times be liquidity constrained, and the absence of liquid interbank markets is a factor here. But it is safe to say that this is not the case when central banks are concerned with excess liquidity. Clearly, the current issue is that banks do not lend out their excess liquidity and not because they might be liquidity constrained.

Taking as a starting point a situation in which banks are not liquidity constrained, what effect would the injection of liquidity have? First, if it is meant to accomplish a monetary easing, it would need to create excess liquidity. This is because if banks were to hold on voluntarily to the extra liquidity, the macroeconomic effects of the injection are likely to be negligible. Next, assuming that the injection of liquidity creates excess liquidity, the key steps of the transmission mechanism are:

  • Open market operations—For simplicity, it is assumed that the central bank injects liquidity via an open market operation; that is, it purchases central bank bills or short-term government securities from market participants. This operation simultaneously injects liquidity and lowers the interest rate on these instruments, which is the hallmark of a successful monetary easing.
  • Reducing the interest rate on short-term securities—This means that these instruments have become less attractive compared to longer-term securities or lending activities (provided that the relative riskiness of these investments has not changed). Moreover, assuming that excess reserves are not remunerated, banks would now be less motivated to hold more reserves that earn them no return. Both factors give banks more incentive to invest in longer-term securities and expand lending.
  • Reducing the lending rate—Assuming a competitive lending market with little friction (for example, banks are well able to judge the creditworthiness of their potential customers), expanding lending will require a reduction in lending rates. Hence, lending volumes should increase and lending rates decline.

The transmission mechanism just outlined corresponds to a standard interest rate transmission channel in which the injection of liquidity leads to a decline in interest rates across maturities and activities and stimulates lending. There are three points to note when applying this transmission mechanism to the Pacific island economies and specifically to the question of the implications of excess reserves at the systemwide level:

  • An increase in lending will not necessarily absorb all excess liquidity—As Garreth Rule noted: “For the system as a whole the quantity of reserves is determined by the balance sheet of the central bank. This means that once reserves are in the system there are few places for them to go. Some could seep into cash in circulation or into government balances. But, if a bank buys an asset (even in foreign currency) or makes a loan then reserves will return to the central bank at another bank” (Rule 2012a, 15). Hence, the existence of excess reserves at the systemwide level does not necessarily imply that the monetary transmission mechanism has been ineffective; rather, changes in lending volumes and interest rates are much more relevant here.
  • The transmission mechanism hinges crucially on the assumption of a competitive lending market with little friction—This is key for interest rate pass-through and for a related rise in lending volumes. However, this assumption may not fully apply to the Pacific, as discussed later. In a nutshell this is because of a low level of competition, high costs of screening customers and loan proposals to identify bankable projects, and the use of lending as a gateway to other profitable banking activities such as foreign exchange trading, under certain circumstances, limiting interest rate pass-through. As a result, lending volumes may not increase by much in response to a monetary easing.
  • Effective interest rate transmission in advanced economies is supported by a number of channels that are mostly absent in the Pacific—Specifically, in advanced economies deep financial markets ensure that short-term interest rates under the control of the central bank have influence over a wide range of interest rates, especially long-term rates. Key elements of the transmission mechanism are expectations over the future path of short-term interest rates—that is, expectations regarding the future monetary policy stance—as well as term and risk premiums. This is an important part of the interest rate channel given that long-term interest rates should be more important for investment, durable consumption, and housing decisions, which are highly interest rate sensitive. In Pacific island economies, underdeveloped financial markets imply that this part of the interest rate channel is likely to be very weak. Likewise, in advanced economies the interest rate channel is reinforced by the effect of interest rates on asset prices, which leads to the asset price channel. In what is known as the balance sheet channel, asset prices also affect the value of collateral and the balance sheet of households and companies more generally, which affects their access to loans. Both channels are likely to be weak in the Pacific islands.

Another question is the source of excess reserves in recent years. A key factor is probably the negative effect of the global financial crisis on aggregate demand due to declining export earnings and general uncertainty about economic prospects. The reduction in aggregate demand would have lowered import demand sufficiently to lead to a buildup of foreign exchange, which did indeed happen in the region. Central banks chose for the most part not to sterilize the foreign exchange buildup—probably in the hope of easing the monetary policy stance and stimulating lending, which would account for the rise in excess reserves. The effect on lending was probably quite limited owing to the weaknesses in the transmission mechanism already outlined, and generally soft demand conditions.

The upshot of this discussion is that concerns about the effectiveness of the monetary transmission mechanism are well justified, but the issue is mainly the interest rate channel. Given the importance of bank lending as the main transmission mechanism, the next section takes a closer look at interest rate pass-through in the banking sector. The other main result is that the existence of excess liquidity in itself is not sufficient to show that monetary policy has been ineffective. To the extent that excess reserves were built up in the few years following the global financial crisis, these can likely be attributed to a rise in foreign exchange inflows that were not sterilized by the central bank. The increase in excess reserves probably did not stimulate lending much but, even if it had, this would have been only a secondary factor for the systemwide level of excess reserves, which is ultimately controlled by central banks and not by the lending decisions of banks.

Interest Rate Pass-Through in the Pacific

The market for the supply and demand of credit is not like a “normal” market in which a large number of firms (banks) intermediate this supply and demand, and low barriers for the entry and exit for firms ensure a high degree of competition. If this were the case, the pass-through of changes in the cost of funds should be immediate and complete. But banking is not a market with low barriers for entry and exit: banking requires trust, because savers need to trust banks to return savings they deposit, and trust is not easily acquired. For the same reason regulatory barriers for entry and exit are high. Hence, competition in the banking sector is limited.

Another characteristic of banking is that financial intermediation is inherently plagued by credit market frictions such as asymmetric information problems and costly contract enforcement. These make lending a costly activity and justify the existence of banks in the first place (Mishra, Montiel, and Spilimbergo 2010). Both factors—limited competition and credit market frictions—tend to reduce interest rate pass-through, and this section looks at the underlying channels.

Limited Competition

Limited competition in the banking sector has an impact on markups, the degree of pass-through, and, in conjunction with adjustment costs, it can make interest rates sticky. We now take a closer look at these impacts:

  • Markup—Imperfect competition in the banking sector implies that individual banks have market power in both the lending and deposit markets. This implies that banks set lending rates as a markup over the marginal cost of loanable funds given by the forgone return on government securities plus marginal intermediation costs. The less competitive the banking sector, the higher the markup (Mishra, Montiel, and Spilimbergo 2010).
  • Pass-through—In a world of perfect competition with complete information, price equals marginal cost and pass-through is immediate; that is, prices adjust instantly and fully to changes in marginal costs.3 As the degree of competition decreases—and leaving aside for the moment the possibility of collusion—the degree of pass-through tends to diminish. For example, for a monopolist facing a linear demand curve, only half of the change in marginal costs is passed through to prices. However, the degree of pass-through does not necessarily decline with the degree of competition; instead, this depends on the functional form of the demand curve.4 Limited competition by itself will not affect the timing of pass-through—that is, to the extent that pass-through occurs—as this is instantaneous. For interest rates to become sticky, another ingredient is needed in the form of adjustment costs.
  • Adjustment costs—Similar to the introduction of sticky prices in New Keynesian macroeconomic models,5 for lending rates to be sticky it is necessary to combine a limited degree of competition with fixed adjustment costs of changing lending rates (Cottarelli and Kourelis 1994). The latter, for instance, could take the form of costs of printing promotional materials to reflect new lending rates, costs of adjustments to the information technology systems, and changing loan manuals. These adjustment costs matter if, for example, the cost of funding falls due to an easing in the monetary policy stance. In this case, banks have to determine whether incurring the adjustment costs of lowering lending rates to pass through lower costs to customers outweighs the cost of not doing so—and therefore risk losing customers, as banks that lower their lending rates gain a price advantage. Lending rates will generally be sticky if banks find it optimal in the presence of adjustment costs to keep lending rates unchanged.

A key factor for the stickiness is the elasticity of loan demand with respect to lending rates. Low elasticity implies that customers are less likely to leave a bank in response to better lending rates on offer elsewhere; in other words, loan customers are relatively insensitive to the lending rates on offer when the loan-demand elasticity is low. In this case, banks have little incentive to reduce lending rates to pass through lower costs of funding after an easing of the monetary policy stance because they run little risk that their customers will go elsewhere. This raises the question of under what circumstances the loan-demand elasticity is low, leaving loan customers insensitive to lending rates. A major factor is a lack of choice: customers may be forced to stick with a bank that does not pass through lower funding costs if they have few alternatives to turn to. Lack of choice is likely to arise from limited competition—there are not many banks to choose from—and from underdeveloped financial markets that offer few alternatives to bank loans as a source of funding. In both situations, lending rates are likely to be sticky.

Short-term elasticities are likely to be lower than long-term elasticities. This is because in the long term it may be possible to access alternative forms of financing such as bond or equity issuance. Hence, lending rates are more likely to be sticky in the short rather than in the long run. Cottarelli and Kourelis (1994, 6) summarize this situation as follows: “The demand for loans of each bank will indeed be less elastic in markets with fewer competitors, barriers to entry, or in the absence of sources of finance alternative to bank loans (e.g., other financial intermediaries, foreign capital markets, commercial paper or bankers’ acceptances markets). In these markets lending rates may show a limited response to changes in money market rates.”

The discussion so far has abstracted from the possibility of collusion. In oligopolistic markets, price stickiness may emerge because maintaining collusion is more difficult when prices change frequently; hence, the colluding banks would tend to keep lending and deposit rates stable (Cottarelli and Kourelis 1994). It should be noted, however, that there is not necessarily a monotonic link between the degree of stickiness and the concentration of the banking sector. What matters is whether oligopolistic collusion arises or not, which can be the case for different degrees of concentration in the banking market. That said, the small number of players in the Pacific banking market may indeed make it easier to maintain an effective degree of collusion. The rapid increase in private sector credit following the entry of BRED Bank shows how a new entry can shake up the market and further its development. Finally, collusion could lead to an asymmetric pass-through in the sense that colluding banks are more likely to pass through an increase in funding costs than a reduction.

A related issue to monopolistic behavior of banks is the presence of switching costs.6 For customers, switching costs come in the form of, for example, costs related to learning about lending rates and conditions at other banks and the paperwork for new loan applications. For banks, switching costs are incurred when they screen new customers. The effect of these switching costs is to segment the market, making it possible for banks to act as quasi-monopolists toward the customers they already have. Given that monopolistic practices can lower the degree of interest rate pass-through, this is a possible side effect of switching costs.

The factors discussed here are especially relevant for the Pacific island economies. Being small inherently limits the size of their banking sector. Besides favoring collusion, the small number of banks operating in these economies tends to lower the elasticities for loan demand, making it more likely that sticky lending rates are an optimal business strategy. A contributing factor is that underdeveloped financial markets mean there are few alternatives to bank loans. Switching costs are another factor relevant for the Pacific because the absence of credit bureaus and limited use of financial tools such as credit cards and bank accounts imply that it is difficult for banks to screen new customers who may not have an extensive financial history that can be easily tracked. Hence, for many customers it may not be easy to obtain loans from other banks, which hardens the quasi-monopolistic behavior of their current bank.

Credit Market Frictions

Asymmetric Information

Asymmetric information in the context of bank lending means that a bank knows less about the projects it has been asked to finance than borrowers do. Specifically, it does not know how risky these projects are. If a bank-financed project fails, the borrower is likely to default, leading to losses for the bank. Charging a high interest rate to compensate for this risk can be a self-defeating strategy: potential borrowers with safe projects that have a relatively low rate of return will withdraw from the market, leaving only those with high-risk projects (adverse selection). Alternatively, borrowers that obtained loans for a relatively safe project may use the money to pursue riskier projects with a higher rate of return instead (moral hazard). In either case, the bank ends up financing a high-risk project instead of the low-risk one it would rather finance. To avoid this, banks can pursue an alternative strategy that Lowe and Rohling (1992, 4) summarize as follows: “Faced with this situation, the bank will elect not to increase its lending rate even if its cost of funds increase. In such an equilibrium, the bank will set the loan rate below the market clearing rate and ration credit. The interest rate will exhibit upward stickiness.”

Major banks operating in most countries tend to be foreign owned, which means their lending rates need to incorporate country risk premiums set by their headquarters. As a result, there is likely to be a lower bound for how much lending rates can fall, making them both upwardly and downwardly sticky.

Critically, credit rationing implies that setting the lending rate and the loan amount are two different decisions. With the lending rate set low enough to make safe but low-return projects viable, there will also be a large number of high-risk projects seeking loans at this rate. Banks will decline to extend loans to high-risk projects to the extent they can identify them, which implies that the decision on whether to finance a project and how much to lend is indeed separate from the interest rate decision. To be able to discriminate against high-risk projects, banks need instruments other than the interest rate; these will likely include an extensive screening of borrowers and collateral requirements as well as limiting bank exposure by requiring borrowers to put equity into their projects.

The need to screen borrowers suggests that the problem of asymmetric information is likely to be more acute in the Pacific than in advanced economies because the necessary instruments, such as credit histories and credit scores, are less available. Hence, it is more difficult (and costly) for banks in the Pacific to determine whether potential borrowers have a track record of implementing low-risk projects and living up to their loan obligations. Other potential remedies such as collateral requirements are hampered by an absence of suitable collateral and asset registries, and costly enforcement of collateral claims. As a result, the strategy outlined of charging relatively low lending rates compared to the risk environment, while rationing credit, is likely to be more prevalent in the Pacific than in advanced economies. This has a number of policy implications:

  • Interest rate pass-through is likely to be limited. This is shown formally in Mishra, Montiel, and Spilimbergo (2010), where the pass-through coefficient is a decreasing function of the slope of the marginal intermediation-cost curve. As they explain: “What this means is that if a deficient institutional environment causes problems of asymmetric information and costly contract enforcement to generate a steeply rising cost of financial intermediation when banks try to expand their lending, banks are less likely to adjust their lending rates in response to changes in the central bank’s policy rate” (Mishra, Montiel, and Spilimbergo 2010, 21).
  • Even if the central bank manages through moral suasion, for example, to lower lending rates, this is unlikely to affect the volume of lending, because the latter is a separate decision. Hence, relatively little is gained from solely lowering lending rates from the viewpoint of providing a monetary stimulus to the economy.
  • In advanced economies, a lowering of policy interest rates tends to boost asset prices and therefore raises the value of collateral and equity, which in turn facilitates bank lending. This is the balance sheet channel already mentioned that reinforces the interest rate channel. In principle, collateral and equity could be especially important for bank lending in the Pacific because these are among the few instruments banks potentially have to discriminate against high-risk projects. Hence, a balance sheet channel could have important effects in the Pacific on the volume of lending, but unfortunately it is likely to be weak, if it exists at all, because of the region’s underdeveloped financial markets.

Relationship Banking

Building a close relationship with customers is a way for banks to overcome the asymmetric information problem. This may give rise to an implicit contract where banks with close ties to their customers offer relatively stable retail interest rates to insulate customers from volatile market rates (Kwapil and Scharler 2006). For risk-averse customers, such interest rate smoothing provides a valuable insurance service, and banks might be willing to provide this if they can charge a higher average rate or see the lending business as a way to open doors to other business transactions.7 In a literature survey for the euro area, where relationship banking is common, Kwapil and Scharler (2006) find that the immediate pass-through from money market rates to retail interest rates is only 0.55 and the long-run pass-through remains mostly below 1. They interpret these results as indicating that banks indeed insulate their customers from volatile money market rates by absorbing part of the changes in money market rates. For the United States, where banking plays a less central role in financial intermediation than in the euro area, they find that interest rate pass-through is higher and mostly complete in the long term.

The Pacific has much more in common with the euro area than the United States in terms of the importance of banking for financial intermediation. With banks in the Pacific dominating financial intermediation, it also makes sense for them to build up close relationships with their customers to overcome the asymmetric information problem. Indeed, doing so may be their best option, given the relative dearth of other tools such as the availability of credit scores or easily enforceable collateral claims. The opportunity to cross-sell other products, especially foreign exchange services, provides banks in the Pacific with another incentive to smooth interest rates to preserve customer relationships.

Conclusion

The concern with excess liquidity significantly reflects a frustration with ineffective monetary policy transmission mechanisms. This chapter shows that the issue is not excess liquidity in itself, but a weak interest rate channel. Before discussing policy implications, it is useful to outline conditions in the aftermath of the global financial crisis as background:

  • The crisis damaged the balance sheets of many international banks, and this typically leads to a curtailment of lending until capital is rebuilt. To the extent that this affected banks operating in the Pacific (ANZ and Westpac, for example), this would have dampened lending activity in the years following the crisis because rebuilding balance sheets often takes considerable time. Businesses and private households that borrowed heavily during the boom years may also need to repair their balance sheets and will curtail borrowing in the interim.
  • Banks became considerably more risk averse following the global financial crisis and tightened lending standards. This change in attitude is likely to have affected all banks operating in the Pacific. For international banks such as ANZ and Westpac it is possible that their headquarters imposed higher country and lending risk premiums on their Pacific operations in the period following the crisis.
  • The global financial crisis led to a reduction in demand and left the economic outlook clouded for years as elements of the crisis such as the European debt situation had not yet been resolved. This reduces private sector demand for credit and raises lending risk.

In sum, without central bank intervention, these factors combine into a tightening of credit conditions, driven both by the supply and demand side. The task for monetary policy is to offset these factors by easing the monetary policy stance; that is, a reduction in policy interest rates. As discussed, central banks in advanced economies can draw on mechanisms such as the asset and bank lending channels to reinforce the effect of lower interest rates. In the Pacific, however, central banks have to rely on the interest rate channel alone without these “boosters.” In addition, the degree of interest rate pass-through is likely to be limited, leaving monetary policy potentially ineffective. And some central banks in the Pacific have also reached the “lower-zero-bound” constraint for policy interest rates that is also hampering advanced economies.

For central banks in the Pacific, a higher degree of interest rate pass-through in the current situation would be desirable to make the monetary transmission mechanism more effective. The type of policy measures—if any—for achieving this will depend on the source of lending rate stickiness and will be affected by:

  • Monopolistic behavior of banks—This chapter has shown that sticky interest rates can emerge naturally from the monopolistic behavior of banks even without collusion, especially if loan demand elasticities are low. If this is the main source of sticky lending rates, the introduction of more competition and regulatory interventions that mitigate monopolistic behavior are the most obvious policy responses, even though these would take effect only over the long term. But is monopolistic behavior really the main source of sticky lending rates? If so, one would expect bank lending to be a profitable business, but Chapter 18 shows that lending operations in the Pacific although profitable are not extraordinarily so. Another argument is that the downward stickiness of lending rates in the aftermath of the global financial crisis reflects collusion. It is, however, also possible that lending rates did not decline by much in the years immediately following the crisis because of higher country and lending risk premiums, and the setting of low lending rates during normal times to avoid adverse selection and moral hazard problems. In both cases, lending rates could not significantly decline because they were already close to their lower bound. Hence, it is by no means clear that the stickiness of lending rates in recent years reflected “untoward” behavior of banks either in the form of collusion or monopolistic behavior.
  • Asymmetric information—From a policy perspective, stickiness of lending rates that stems from asymmetric information has two implications. First, as noted, lending rates may have reached their lower bound already as they were set low to begin with. A further reduction in lending rates would only be possible if risk premiums decline, but this is outside the control of central banks. Second, even if it were possible to lower lending rates through administrative measures or moral suasion, this would not increase the amount of lending because setting interest rates and determining the amount of lending are two separate decisions. The only promising policy intervention is to improve the functioning of the banking sector by making it easier to expand lending through lowering the cost for screening borrowers. That is, the asymmetric information problem itself needs to be addressed through the introduction of credit bureaus and other institutions that facilitate bank lending. Obviously, this can be successful only in the long term, but it would have benefits that go well beyond improving the effectiveness of the interest rate channel by making the advantages of financial intermediation available to a larger part of the population.
  • Relationship banking—The implications here are similar to those for asymmetric information in that lowering lending rates through administrative measures or moral suasion would not necessarily achieve an increase in the amount of lending. Lower “headline” lending rates achieved by such an intervention would not automatically benefit customers that are in a close relationship with their banks. This is because they would not expect to benefit from the best rates during periods of low interest rates, just as they would not expect to be charged the highest rates during other times. The policy options in this case are very limited.

The bottom line is that there is very little that can be done in the short term to improve the functioning of the interest rate channel in Pacific island economies. However, in the long term, policy options are available to improve considerably the functioning of the banking sector, especially if it is currently hampered by monopolistic behavior or asymmetric information.

References

    CottarelliCarlo and AngelikiKourelis.1994. “Financial Structure, Bank Lending Rates, and the Transmission Mechanism of Monetary Policy.IMF Working Paper 94/39International Monetary FundWashington.

    • Search Google Scholar
    • Export Citation

    KwapilClaudia and JohannScharler.2006. “Limited Pass-Through from Policy to Retail Interest Rates: Empirical Evidence and Macroeconomic Implications.Monetary Policy & the Economy Quarter 4: 2636.

    • Search Google Scholar
    • Export Citation

    LowePhilip and ThomasRohling.1992. “Loan Rate Stickiness: Theory and Evidence.Research Discussion Paper 9206Reserve Bank of AustraliaSydney.

    • Search Google Scholar
    • Export Citation

    MishraPrachiPeter J.Montiel and AntonioSpilimbergo.2010. “Monetary Transmission in Low Income Countries.IMF Working Paper 10/223International Monetary FundWashington.

    • Search Google Scholar
    • Export Citation

    RuleGarreth.2012a. “Introduction to the Central Bank’s Balance Sheet.CCBS-PFTAC course on Monetar Transmission Channels Liquidity Conditions and Determinants of InflationBank of EnglandLondon.

    • Search Google Scholar
    • Export Citation

    RuleGarreth.2012b. “The Impact of Surplus Liquidity.CCBS-PFTAC Course on Monetary Transmission Channels Liquidity Conditions and Determinants of InflationBank of EnglandLondon.

    • Search Google Scholar
    • Export Citation

This chapter is based on discussions at and subsequent to the Sixth Seminar on Monetary Transmission Channels, Liquidity Conditions, and Determinants of Inflation held at Honiara, Solomon Islands, July 18–23, 2012. Given the informal nature of the seminar, no specific attributions are made, but the central contribution of Garreth Rule (Bank of England’s Centre for Central Banking Studies) should nevertheless be highlighted. Thanks are also due to Ole Rummel (Centre for Central Banking Studies), Adam Gorajek (National Reserve Bank of Tonga), and John Vaught (Pacific Financial Technical Assistance Centre financial sector supervision advisor at the time of the seminar) for their valuable contributions.

1

The discussion draws on Rule (2012a).

2

The following discussion is based on Rule (2012b).

3

This discussion draws on Lowe and Rohling (1992).

4

For a detailed discussion, see Cottarelli and Kourelis (1994).

5

These stem from the assumption of monopolistic competition combined with so-called menu costs that represent price adjustment costs.

6

This discussion draws on Lowe and Rohling (1992).

7

For the risk-sharing argument, see Lowe and Rohling (1992).

    Other Resources Citing This Publication