Thursday, August 5, 2010


Its official. The first phase of Quantitative Easing did not really work out all that well. And its almost certain that on August 10, the Federal Reserve will decide to take another shot at it.

But apart from inscrutable nomenclature, which essentially means monetization or the creation of money by a monetary authority to acquire assets from either the banking system or the government, there is a lot else that is wrong with it. Traditionally, monetization has almost always been undertaken to allow a government to spend more than it earns and it usually takes the form of the monetary authority paying the government for debt instruments issued by it. This debt can take many forms, from low coupon instruments without any defined repayment which the government issues only to the monetary authority (who else would buy them?) to regular government bonds which the monetary authority could actually sell to market participants if times turned. In all these cases, the monetary authority has an asset it can value at par, which allowed it to match its assets and liabilities without booking a revenue charge. But that makes monetization just another form of government borrowing, right? Well no, it doesn’t.

If one looks beyond the veil and consolidates the account of the monetary authority with the government, then debt owed by the government to the monetary authority would disappear from the books leaving only the money issued as a liability on the consolidated account. This would effectively mean that all money backed by domestic government debt, is in essence, created without any incremental assets to back it. So why the elaborate farce? Well, it sounds so much better when you hear that the currency you hold is a liability of the monetary authority fully backed by assets, doesn’t it?

But the Fed has taken this travesty to a completely different level. It has bought, and wants to buy mortgage backed securities (MBS) from the banking system hoping that this will encourage banks to start lending again. There is also the highly desired possibility that it will result in US Dollar weakness and higher inflationary expectations. The Fed hopes that low interest rates and elevated inflationary expectations would boost aggregate demand and allow business to recover. Lets count the ways that this can go wrong.

1) These securities are liquidity impaired for a reason. They are failed contracts which would trade at a steep discount to par value if they could be traded at all. The reasons why they aren’t traded include the lack of confidence buyers have in paying anything for them and the fact that if they were sold at market clearing rates, most banks in the US would be bankrupt. In buying these, the Fed is committing to hold them to maturity or default, whichever happens earlier as no bank will ever buy them from the Fed. At the moment, banks are valuing these securities at close to their par value and the only way they will not go bankrupt will be for the Fed to buy them at this value. This means the Fed would be left holding securities worth less than what they paid. Its the same as issuing currency notes at a discount.

2) Moreover, since these aren’t traded and there is no market valuation available, the Fed will also value them at cost, effectively delaying loss recognition to when defaults occur. Its criminal when a commercial enterprise cooks its books in this manner, but for the monetary authority of the world’s largest economy, its despicable. I am assuming defaults will occur because the manner in which these securities were structured and issued ensured that many are reasonably suspect and it would be naive to believe that banks would sell only the one’s that aren’t.

3) The US Government has been extremely critical of China’s monetary policy which allows the Chinese monetary authority to create money by buying foreign currencies. They believe that it distorts the currency market and allows China to weaken its currency below levels justified by its fundamentals. But by issuing currency in exchange for impaired domestic assets, the US wants to achieve the same outcome. Initial indications of currency weakness already exist with the US Dollar hitting a 15-year low against the Yen on mere expectation. Why doesn’t the Fed just jump on the bandwagon and buy Euros and Yen instead. Then if the US Dollar depreciates, the Fed will actually be in the money. The only catch in doing so, apart from the loss of face internationally, would be that banks would continue to hold these impaired securities along with the potential loss. Which makes me wonder, does the Fed really want to undertake quantitative easing to improve the economy or just nationalize bank losses in the guise of QE?

4) Another assumption the Fed is making is that consumers and businesses would want to borrow. But businesses are hoarding over $1.24 tln in cash and any expenditure would come from this first. US Household Debt still totals $2.5 tln with the average household spending over 17.5% of disposable income on financial obligations. There is a significant possibility that US businesses may not want to invest, and US households may not want to borrow more despite low rates and easy availability. Think about it, residential mortgage rates are at an all time low. Does this sound like borrowers want to borrow, but lenders don’t have the cash? I would bet, like the previous round of quantitative easing, banks will park this new money in reserves with the Fed, and the Fed will push the string harder with another round of monetization.

QE III, anyone?


  1. we are no longer under the dollar standard.we are under the MBS(AAA+ rated by Moody's) standard.the assets that back the dollar are toxic pieces of paper.we should all subsidise Jose in california so that he can buy his mcmansion and keep his payments going so that the MBS remains above water.
    btw, is the mark to market accounting still suspended for such securities?

  2. the us mortgage securities for one?
    also,i thought the european bank stress tests passed because of suspending m2m. right?

  3. US accounting practice, since the change in 2009, does not recognize the current market for MBSs as orderly. Hence, my earlier response. The US FASB did not suspend mark-to-market requirements; just eased them. In the absence of orderly market conditions, banks have to quantify the difference between cost and market value, but account for/write off only the portion they believe is caused by "estimated future credit losses". All other aspects of market pricing like liquidity etc. need to be mentioned below the line only. Since estimating future credit losses are decided by each bank, at its discretion, the result is the same as a suspension of MTM requirements without an actually suspension. Financial innovation at its best, isn't it?

    For stress tests, MTM is typically ignored as it can give both, pessimistic & optimistic biases to the results. Assets are taken at par value and a haircut decided by the authority conducting the stress test is applied. Typically, these haircuts will be the relevant authority's "worst case" estimates for different kinds of securities. It is possible that these "worst case" estimates may have been skewed by the ECB to ensure that banks survive it, but that can only be conjecture.


  4. thank you. your blog is an excellent read.after years of reading the western analysts and economic history of the west thru both socialist and hayek/misesian terms, its finally good to see how things pan out at home.not sure if it is your forte,but one day,perhaps you could address why sound money advocates are so few in india and why deficit financing/monetizing continues unprotested.

    also,not connected to this post directly,but i read in one of your posts that if we must base the rupee value(vs dollar) using REER, then we should go back too the 1988 value of 13rs instead of the 1993 value of 40.could you clarify why 13 is a good point to start?

  5. Commenting on why there aren't many sound money advocates isn't my forte at all, but I have a theory about it which i will elaborate in a post soon.

    About the rupee, well, its like this. When Rajiv Gandhi came to power, he wanted to take India to the 21st century. His government allowed industry to import capital goods on the condition that a similar amount of products will be exported by such companies in 7/10 years. This was just one of the measures which were introduced and all of these together resulted in a foreign exchange flow mismatch, which was compensated by sovereign foreign exchange borrowings. Starting 1988, foreign lenders started getting a bit nervous about our ability to repay and financing the trade deficit became increasingly difficult till we hit the wall in 1991. Since 1988 however, the rupee had started depreciating, a process which peaked in 1991 and the loss of value in this period wasn't because of inflation differentials, but a demand - supply mismatch.

    REER, as a model, is an incremental Purchasing Power Parity(PPP) adjustment model. It uses a base value and then adjusts it for inflation differentials in an attempt to align PPP dynamically. But in doing so it assumes that PPP for the base year was accurate to begin with.

    When we take any year after 1991 as a base for REER, we take a value which includes depreciation/devaluation caused by the demand supply mismatch, which had absolutely nothing to do with PPP. The demand supply problem has since been solved, pretty comprehensively at that, with foreign exchange reserves now far in excess of what may be desirable from a trade perspective. Hence, it is essential that we reverse that portion of the rupee decline that was caused by shortages, at the very least. This can only be achieved by taking any year before 1988 as the base, calculating inflation differentials from then and arriving at a rupee value that will be an accurate representation of its PPP value.

    The move from the current exchange rate to this PPP rate cannot be undertaken in one shot. It would cause mayhem. But it will cause a shift in perspective, from weakening to strengthening, which is essential to India's future status as one of the largest economies in the world. India's PPP GDP is about 2.3 times the exchange rate GDP, based on the CIA factbook. Taking this as accurate, the PPP exchange rate would be 46/2.3, or Rs. 20/USD.