Tuesday, April 19, 2011

Shut-up, CNBC: Carney on Muni bonds edition

Yesterday, Carney, when discussing risks to the Municipal markets said something I feel is inaccurate and deserves correction.

5. Information Cost is High. Muni issuers are not subject to the same disclosure requirements as corporate borrowers. The market is illiquid so pricing is opaque. The swaps market—the market for tradable credit protection—is thin and unreliable. This means that bond buyers may be taking on risks that they are not aware of. This is a recipe for panic once a triggering event occurs. 
First of all, CDS trading on a LOT of things is very thin. The municipal market is not a monolith, it is composed of thousands of issuers and so, yeah, swaps are probably pretty thin for many issuers. Just like bond and CDS mkts would be thin for most corporate issuers. Second of all, municipal market is dominated by retail, for obvious reasons. Joe Smith looking for safe tax-free bonds isn't exactly your average CDS trader.

Secondly, information is not that hard to come by if you know how to look and are not lazy. Anyone that's lending anyone money should do their homework, or at least pay an adviser or fund manager that has fiduciary duty to do it for them. MSRB is a wonderful resource, including a freely available trade history and, as Bond Girl has pointed out before, MSRB's EMMA allows you to find many issues' official disclosure documents.

Lastly, if you are the kind of investor that doesn't need an adviser or fund manager and is buying large amounts of municipal bonds, you can probably afford a subscription to The Bond Buyer and professional research from the rating agencies (ignore the rating and read the analyst's report).
6. Arbitrage Buying Leads to Bubbles. Much of the demand for muni bonds is not a function of credit analysis or a desire for exposure to the revenue streams of local governments. It is done for a technical, legal reason—to take advantage of the tax-free status of muni bond income. This creates an artificially high demand—a bubble—much like Basel accord capital requirements led banks to overinvest in mortgage bonds.
As Bond Girl wrote a few weeks ago, the last few years have seen some demand destruction, not only from a shift in investor's appetite for municipals relative to other securities, but by the departure of a number of leveraged actors that, through the use of short-term funding schemes created additional demand at the long ends of the curve. We're also seeing that "the muni market is transitioning from an interest rate space to a credit space." I can't go into it here, but you should really read Bond Girl's piece, it is excellently written and informative in a way no newspaper ever will be. But, getting back to my point, tax-free income doesn't necessarily increase demand.

Tax-free yield is attractive and investors are usually willing to take smaller yields for tax-free debt because it is still attractive in a taxable-equivalent basis. However, tax-free yields are only attractive to investors that benefit from preferential tax-treatment, limiting the universe of potential buyers--this is why the BAB program was introduced, to stabilize demand by introducing new participants. In my opinion, the tax-exempt yield of municipals actually hurts demand by restricting the universe of potential buyers. One only needs to look at the disparities between non-taxable TEY and BAB yields earlier in the year to see this at work.

The solution here, in my opinion, is for the federal government to refund issuers directly, like in the BAB program. Investors will receive higher yields that are equivalent on a taxable-equivalent basis and municipalities can offset the additional cost of debt service with refunds from the tax collectors paid for by the tax collected on the new, taxable issues. Otherwise, we risk a market where municipals can actually pay a premium, as they trade not only on their taxable-equivalent basis, but also on any extra risk or liquidity premium required by the restricted universe of buyers, increasing volatility and therefore the cost to issuers.

Monday, April 11, 2011

How China's Negative Real Rates Depress Consumption

If you've ever caught me ranting about China on twitter, you've seen me carry on about how negative real deposit rates are an implicit transfer of wealth from households to government. You may also recognize that point from Michael Pettis' China Financial Markets blog. In this post I'm going to try to explain how this transfer works. It's not complicated, but if you don't understand how developing economies differ from economies like that of the US it can be hard to see the mechanism at work.

The basics:
  • Outside the upper-middle and upper classes, consumer credit is not easily available in developing economies. You can't just call Experian and check someone's FICO. Large amounts of the population is unbanked, underbanked or has no credit history at all.
  • The deposit and lending rate (and therefore the spread between them) are set by the central government.
  • Michael Pettis has estimated real deposit rates are suppressed by "at least 400-600 basis points" (China Financial Markets)
Negative real deposit rates are a transfer of wealth from depositors and creditors to debtors
Because of the limited access to credit that households have, households are the main source of deposits in the system. Like in other developing and under-banked economies with limited access to consumer credit, in China you need to save money until you have the full price to pay for X good (e.g. durables) which, when combined with inflation, forces the households to accept negative real rates. Accepting a 2% yield when there's 5% inflation may mean -3% real rate, but it's better than the -5% cash yields. Reasons for savings include emergencies, possible medical expenses, savings for a home, vehicle purchase or a child's education or every-day cash management. Remember, the rest of the world doesn't use their Capital One to pay for their groceries. With this kind of saving pattern, real rates have inverse effects on saving because, the more negative real rates are, the higher the savings must be to achieve a savings goal or maintain the real value of the savings balance. Without access to credit, negative real deposit rates force the household sector to save more. Money channeled towards savings by the household sector is money that is not spent on consumption. An approximation of total tax on households--and consumption--would be the product of the average daily balance of total deposits multiplied by times the gap between market and government-set rates multiplied by the percentage of household deposits in the system. That estimate excludes any effect from misallocation of resources by borrowers.

Who are the creditors? Who is receiving the transfer?
In April 2009, the Hong Kong Institute for Monetary Research published a paper which claims that state-owned enterprises (SOEs)--which account for about 37.6% of total value added in their respective industries and 25% of GDP--are only, or mostly, profitable due to a preferential cost of capital. According to the authors, "SOEs’ profits would have been entirely wiped out if SOEs were made to pay the same interest rates as otherwise equivalent private enterprises." How big is the problem? Well, "although SOEs’ contribution to the Chinese GDP was around 25%, they received about 65% of total loans."

Seeing as how a large portion of the SOE are involved in investment-related activities (in the GDP sense) and SOE's account for a majority of Chinese GDP the easy conclusion is that investment is being financed by taxing households through the banking sector. Net-winners? Anyone involved in that supply chain and the SOEs. So the transfer is moving from Households to government and business. Which businesses? Well, since, "about 41% of private enterprises have no access to credit and 56% have no access to bank credit," I'm going to guess big businesses.

Negative real rates impede a growth in household consumption by transferring wealth to government and business as long as households bear the cost of those negative real rates.

This leaves us with one way to increase consumption:
  1. reduce the burden carried by households as a result of negative real rates.
And two possible ways of achieving that:
  1. Raise real rates
  2. Transfer some of the cost of negative real rates elsewhere
Barring a flood of foreign depositors itching to deposit funds into RMB-denominated accounts at negative real rates or holding RMB as FX reserves, option two means either government or business. Analyzing the effect to the SOEs is beyond the scope of this post, but since profits ultimately go back to the state, we can look at SOEs and government as one, and consider that subsidy as a tax. It's simply a way for the government to decide how citizens spend their own money. Unless the government stops trying to do that (fat chance), reduced subsidies to SOEs and government would just require more taxes/state-borrowing or less spending, of which the ultimate recipients are the household sector again.

Essentially, the other two beneficiaries of negative real rates, households and private industry with access to credit, are free-riding on this policy and benefiting from low-cost financing, creating a regressive re-distribution of household wealth. This is one of the many reasons we've seen restrictions, like higher down-payments, placed on mortgages of second and third home purchases.

The simplest fix to the problems caused by negative real rates (under-consumption, regressive redistribution, resource misallocation, asset-price inflation) is simply to raise real rates. Of course, raising real rates could cause a surge in NPLs from SOEs that wouldn't be profitable without the implicit subsidy. In the event of a SOE being rendered unprofitable by having to access capital at market rates, the implicit subsidy could simply be turned into an explicit one if the firm's activities were deemed important enough. Otherwise, the firm would go away or shrink, eliminating the dead-weight loss from misallocated resources. The banking system? The risk is already implicitly (and sometimes explicitly) socialized and the problem won't simply go away if Beijing keeps waiting.

Moving real interest rates up doesn't necessarily have to be contractionary for the economy. Sure, the loss of the implicit transfer from households to SOEs / Government would reduce investment capacity, but at the same time it would increase consumption capacity by an equal amount, although probably not the kind of consumption Beijing would prefer. Given the reduction in losses from negative real rates, households would be more able to absorb tax increases if Beijing wished to keep subsidizing investment.

Additionally, stepping closer towards market rates would allow the expansion of consumer credit. Widespread access to consumer credit wouldn't really co-exist well with real negative rates unless there was exterior financing, which would require major changes to the current account. Notwithstanding Q1 2010 numbers, I simply don't see consistent trade deficits for China in the horizon, so the next the easiest and healthiest road to increasing access to consumer credit is simply positive real rates.

Why increase access to consumer credit? In many ways--although not all--savings can be replaced by access to credit. Many people have savings so that they have enough purchasing capacity in short notice in case of an emergency. Credit can replace that cushion in many cases, reducing the need for short-term time or demand or deposits. Following this logic, giving someone access to credit allows them to spend deposits on consumption because the emergency purchasing power they needed is still there in the form of credit. Just this action, without any actual household borrowing, would shift some savings to consumption by reducing the quantity of household savings that need to be parked at deposit institutions.