MB has hosted many debates around cultural differences between countries and groups of people. But I’ve never understood what the writers are referring to when they use the term cultural to explain economic differences between, say, countries. As research certainly supports the notion that with a given set of choices generally we act alike. In this sense Germans are Greeks and vice versa.
Wikipedia defines one of the meanings of the term “cultural” to be, “The set of shared attitudes, values, goals, and practices that characterizes an institution, organization, or group”. Is that what the writers mean when the term cultural is used? If so then these are things that are created by humans and are not inherent differences in the human condition. Germans are not Greeks because of systems created by Germans.
I do greatly respect the work of John List Professor of Economics at the University of Chicago who “focuses on using economic theory and empirical tools to further economic science”. His team’s great body of research reveals to me that in given circumstances people of a great variety of types react similarly, although some differences do occur with reactions and choices between students and professionals.
List’s type of research is heartening if we are to try and have global commonality of regulation governing financial markets because if Americans can’t become Aussies what would be the point? There are very good reasons to have commonality of global financial regulation but as I’ve argued often, that regulation must be about creating the infrastructure in which markets operate, it cannot and must not be simple rule based which is what the rent seekers seek! It’s also why bad regulation has had some very bad unintended consequences.
I am going to outline two examples of bad regulations with almost butterfly wing effects that demonstrate the decimation that bad regulation can have. One historical, the other example we are sadly living through at the moment.
To demonstrate my point I’m firstly going to use the analogy of a roulette wheel in a make believe casino, which has been built with 2 number 36s on it. No one could see the numbers on the wheel, and the winning number was simply electronically displayed. The casino was regulated with only a few designated firms appointed by regulation to provide a certificate containing their opinion that the roulette wheel was in AAA condition.
The punters start playing roulette with a few noticing that 36 seemed to be winning more than expected. It took awhile to work this out but once known the punters in the know piled into 36 and other more likely winning combinations. The croupier and his bosses noticed the distortion but with that certificate backing up the wheel’s authentification there was nothing really they could do. In fact to go against the certificate would have been against the law. Now I could write extending the analogy with an extensive scenario of how the punters in the know kept the game going as long as possible. The number one requirement would be the continual entry into the game of punters not in the know ie there has to be losers disguising the activity of the winners. But what do you think would happen in the circumstance described and who’s really responsible?
Now let’s look at the what’s generally accepted as the catalyst for the 2008 financial crisis. “Unscrupulous” brokers and lenders granted millions of mortgages in the US to under qualified borrowers. Why and how could this happen? The answer is actually relatively simple. The casino’s risks were distorted.
Securitisation was the main financial technique used to create the funds to provide most of the mortgages in the US and regulated firms ie credit rating agencies were mandated with the responsibility to give opinions about the quality of debt generated by securitisation of mortgages. However, not only were these rating opinions opaque there were actually faults in their models, which was difficult if not impossible for anyone to ascertain. Besides how rating agencies assess serviceability by borrowers the single biggest error in their assessment was to do with the loan to valuation ratio. Simply once the LVR for a borrower was above 95% the risk assessment flattened out. Two of the three main rating agencies provided a 4.5 multiplier to the risk of a mortgage at 95% LVR and then flattened the risk assessment so that a 110% LVR mortgage was assessed at the same risk as 95% LVR all else being equal.
Therefore lenders were not effectively penalized for what is clearly an increase in risk and therefore were incentivized to lend at ever increasing LVRs above 95%. Whilst it took many years for this incentive to work through the system it certainly did. History shows that the so-called unscrupulous brokers and lenders pushed the system to the limit until it broke. But what history also showed was until the system broke the credit rating agency risk assessment was arguably correct. But it was rising house prices supporting the good statistical performance, which disguised the real risk. In this casino, who’s responsible for the distortion and failure?
Now lets turn to Australia’s own regulated casino, the residential mortgage lending system under which our four main banks operate.
As I’ve written about on previous occasions the amount of capital needed to be allocated against residential mortgages is critical to a bank’s return on capital but more importantly the banks ability to continually generate further borrowings to lend to an ever growing mortgage market. So how this capital is calculated is a very important question as these banks allocate less than 2% capital against their mortgage risk?
The major banks all have internal risk based models under Basel II, which are used to calculate risk weighted assets and therefore minimum capital requirements. Models are built in accordance with the rules outlined by the Basel committee, which have been adopted by APRA. For those so inclined please see here for the full rule set. For this post and at the risk of seeming to over simplify the requirements, I am going to try to only point out the most important regulatory requirement I know that has distorted Australia’s financial system and effects all our lives.
The calculations of risk weighted assets under Basel II models must built around probability of default, loss given default, exposure at default and expected losses with these values determined by fact and historical performance statistics. These broad factors and requirements are meant to apply to all loan assets. Sounds like a sound basis, doesn’t it? Well it’s not and has lead us into a situation similar to the distorted casino described and the US sub-prime securitisation debacle.
There is a fundamental flaw with using the same Basel II model requirements for unsecured retail loans like credit cards and mortgages. To be fair this is recognized to some extent by Basel II rules but seems to be ignored in practice. Performance of credit cards is not directly supported by asset growth, mortgages are.
The primary driver of all Basel II models is probability of default (PD). I’m only giving one reference point from Basel II rules but please research in detail if you want to satisfy yourself:
Retail exposures 252. For retail exposures, banks must provide their own estimates of PD, LGD and EAD. There is no distinction between a foundation and advanced approach for this asset class.
So how are PD and LGD determined in the models of the major banks? Well, we know because of the very big deal the banks make about it, that APRA requires a minimum of 20% LGD. Although this 20% seems arbitrary and conservative, it’s had the opposite effect. The banks are very incentivized to provide low PDs.
The Pillar 3 disclosures of our major banks, although overall inadequate, do make it clear that PD for mortgages is determined by LVR band and historical performance, which in practice is historical default information over the last 5 years. So capital is determined by what has happened yesterday not what may or is likely to happen in the future.
In effect the banks are incentivized to pump credit growth and inflate a property bubble, because as this process occurs performance will be good as defaults and losses are masked by house price growth. Lower capital requirements and a greater ability to raise funds from offshore to pump the housing credit to the detriment of other asset classes reward banks. If that’s the system, then on an individual basis it’s very hard to blame the bankers for gaming to their benefit. After all its government sanctioned just like the credit rating agencies.
Note: The US did not adopt Basel II; the UK did and still does even for banks owned by the government.