Sunday, April 13, 2014

Universities prepare students for society, not necessarily jobs


This morning, I have an oped in the Boulder newspaper, The Daily Camera.  

In the article I argue that recent criticisms that universities ill prepare students for employment and many can't find jobs despite substantial loan debt are misplaced.  Ensuring students are better positioned for employment is in general, not a goal of universities or the federal student loan program.

Instead of criticizing universities for poor performance against inappropriate criteria, government and universities need to better manage expectations for the role of higher education in society.

You can read the full article here.

Thursday, April 10, 2014

The cost of insurance does not control building activity in Florida


Recently The Economist had a couple of articles focusing on Miami.  The above image comes from an article about Miami's recovery from the most recent real estate slump.  The Economist argues that Miami attracts investment from the ultra rich around the world but most notably those in South America.  Compared to apartment prices around the world, Miami is still relatively cheap.

There is a common belief that pricing insurance appropriately will control overdevelopment or reckless development.  Yet, trends in real estate sales and development do not suggest that this belief is true.

The cost of windstorm insurance in Florida has been relatively high since 2005.  The graph above shows that real estate sales have risen since 2008.  

Only recently, has the industry and decision makers expected that policyholders's costs will level or decrease as a result of increasing competition within the industry.  The graph above shows that real estate sales have tapered off since 2011.

It appears that building activity in South Florida is influenced by a number of factors such as, capital market conditions and political conditions in other countries.  But it seems to have little to do with social conditions in Florida including the public's ability to afford the cost of insurance.

Wednesday, April 2, 2014

Why US disaster losses are so expensive...er, cheap??


Sometimes when people lament the financial cost of disasters in the US, I am struck by what seems to me a bargain deal.  There are individual people in this country that have an estimated net worth equivalent to the cost of many of our disasters.  

For example, Hurricane Katrina cost somewhere over $100 billion.  To be sure, this is a lot of money, but this amount includes the destruction of a good portion of a large metropolitan US port city, New Orleans, and destruction in several other cities and states.  That we can obliterate parts of cities in a nation worth $16 trillion and end up with a bill of around $100 billion of which private industry pays a good portion of- this sounds like a basement bargain deal to me. 




The above image helps explain why our nation's disaster losses can get so big and at the same time can also be pretty cheap all things considered.  

The image is from FEMA's National Flood Insurance Program (NFIP) 2013 budget justification.  Despite the title (though more on that in a bit), the image helps explain why flood loss in the US can be so expensive though perhaps cheap in the long run.

"Risk" has many dimensions social, economic and political.  But for the purposes of insurance, "risk" is a measurable uncertainty estimated by analysis of the frequency and severity of a given type of loss event.  Frequent losses of relatively small severity, such as auto accidents are ideal for insurance.  Insurance has greater difficulty managing infrequent and very large losses.    

Throughout history the US has invested heavily in flood control measures such as damns, levees, zoning, etc.  As a result, more of the nation's land became productive- people could build homes, develop business and otherwise conduct economic activity in areas that without flood protection would too frequently experience flooding to make it worthwhile investing in the land.  

However, while investments helped reduce the frequency of flooding it increased the severity of flood events when they occurred.  This phenomena is often called the "levee effect" with credit to the legendary natural hazards expert Gilbert White

As a result, perhaps all of US flood risk is now tail risk- infrequent, large loss. This is the type of loss, as indicated by the graph, that we use the NFIP to manage.  

Managing this type of risk with insurance is clumsy because by conventional insurance practices, it is very expensive.  Using a public insurance program like the NFIP helps us manage the cost of coverage and pay losses over time.  But this also means that the NFIP is not conventional insurance it's a residual market.

There is a general belief that in the long run, the US economy is better off experiencing infrequent large losses than frequent smaller losses because in between large losses the nation prospers through use of the otherwise hazardous land.  In recent years though, there is some speculation that the flood losses have become so big that, in fact, we lose more than we gain.

Deciding which conclusion is true will tell us if our process of flood management works to "buy down" losses as the graph's title suggests.  But does it really matter?  

The nation is fortunate to have a vast network of rivers, coastline and estuaries and a climate that provides abundant rain in most places.  This also means though that few of us don't face some sort of flood risk.  It is hard to imagine large cities and populations up and moving elsewhere so that we can remove flood infrastructure and try to go back to frequent small loss events. It is equally difficult to imagine revolutionary mitigation to homes while the public still feels that they are struggling to keep up with the cost of living. 

Expecting maintenance of flood infrastructure and perhaps some improvements here, perhaps US flood loss is as cheap as we can expect.    

Thursday, March 6, 2014

Beyond McCarran–Ferguson


In January, I attended an insurance conference in Florida where I had the lucky opportunity to pose a question to Florida's Insurance Commissioner Kevin McCarty and Representative Bryan Nelson.  I asked, "What are some of the reasons that proposals for regional and national catastrophe insurance pools fail?"

The question was first met by silence and then very quick agreement between the two men that the policies fail due to a lack of public political will.  As one of them explained (I forget which one and I paraphrase), "It is hard enough to get northern Louisiana to share risk with Southern Louisiana, let alone multiple states to share risk."

If there is one thing that I have learned from studying political science, it is that when the public is blamed for policy failure there is a far more interesting story to be found in the details.  In any case, when a republican politician and an insurance regulator emphatically and quickly agree on something, one should get a second opinion.

Inspired by the response I received, I became curious about why the states' regulate insurance in the first place.  In general, this question is often met with the brief response that the McCarran-Ferguson Act of 1945 (MFA) explicitly gives the states authority to regulate the insurance industry and so, they do.  But this does not sufficiently explain the reasoning behind the "Why?"  MFA simply provides the legal authority for the states to do so.

The question is worthwhile asking because throughout American history numerous instances demonstrate presidents and policy makers seeking Federal regulation of the insurance industry.     One of the continual proponents of Federal regulation has been the insurance industry itself, at least at one point,
"consider[ing] it more advantageous to be regulated by a toothless, laissez-farish mastiff like the Federal Government than by those smaller but possibly more harassing watch dogs, the individual states” (quoted in Alt 2010).  
The MFA was born from a need to clearly designate regulating power after the Supreme Court ruling on the United States v. South-Eastern Underwriters  Association (hereafter as USSEA).  The case challenged a long held precedent that the business of insurance is not commerce and therefore exempt from federal anti-trust regulation and any regulating authority designated to Congress through the Constitutions' commerce clause.

The Supreme Court engrained the precedent in 1869 with the ruling in Paul v. Virginia (PV)- only a few years after the end of the American Civil War.  The contentions bringing about the Civil War were evident in the PV case as the two issues of focus were: 1) the right of citizens of one state to the "privileges and immunities of citizens in the several states" and Congressional power "to regulate commerce with foreign nations, and among the several States."  Moreover, the case involved a New York insurance agent (Yankee) attempting to do business in Virginia (Confederate).

The practicality of the dispute was regulatory authority and taxing ability over the insurance industry.

Justice Stephen Johnson Field, undoubtably aware of the sensitive nature over state's rights that remained after the war, seemed to find a balance between upholding the, then, delicate ideas about freedom and state authority over their respective economies.  Justice Field argued extensively for citizens' right to move about the nation, acquire and enjoy property and pursue happiness,
Indeed, without some provision of the kind removing from the citizens of each State the disabilities of alienage in the other States, and giving them equality of privilege with citizens of those States, the Republic would have constituted little more than a league of States; it would not have constituted the Union which now exists. 
Yet, the extension of this right to corporations made up of citizens was marred by complications, particularly, corporations' ability to amass wealth and power,
The corporation being the mere creation of local law, can have no legal existence beyond the limits of the sovereignty where created. ... Having no absolute right of recognition in other States, but depending for such recognition and the enforcement of its contracts upon their assent, it follows as a matter of course that such assent may be granted upon such terms and conditions as those States may think proper to impose....The whole matter rests in their discretion. 
If, on the other hand, the provision of the Constitution could be construed to secure to citizens of each State in other States the peculiar privileges conferred by their laws, an extraterritorial operation would be given to local legislation utterly destructive of the independence and the harmony of the States. At the present day, corporations are multiplied to an almost indefinite extent. There is scarcely a business pursued requiring the expenditure of large capital, or the union of large numbers, that is not carried on by corporations. It is not too much to say that the wealth and business of the country are to a great extent controlled by them. And if, when composed of citizens of one State, their corporate powers and franchises could be exercised in other States without restriction, it is easy to see that, with the advantages thus possessed, the most important business of those States would soon pass into their hands. The principal business of every State would, in fact, be controlled by corporations created by other States.
Finally, Justice Field argued that regardless of the whole discussion of citizenry and freedom, there is nothing inherently keeping Congress from regulating insurance if it were that insurance was commerce which Field argued quite clearly that it is not,  
Issuing a policy of insurance is not a transaction of commerce
When the Sherman Act came about in 1890, seemingly everyone assumed insurance to be exempt from that, too.

Shortly after the PV ruling, state insurance commissioners created the  National Association of Insurance Commissioners (NAIC) in response to a felt need for some type of organization and consistency of regulation amongst the states.  

With the Great Depression, the federal government became fervent investigators of potential monopolies and their prevention.  According to Alt (2010), research by the federal government found that states were having trouble regulating the industry because of business was concentrated in just a few companies.  

In 1942, the US Justice Department filed suit against the South- Eastern Underwriters Association charging them with fixing premiums and boycotts.      

Justice Hugo Black gave the majority opinion on the case in which he argued that the discussion of
insurance being or not being an act of commerce was ridiculous,
To hold that the word "commerce," as used in the Commerce Clause, does not include a business such as insurance would do just that. Whatever other meanings "commerce" may have included in 1787, the dictionaries, encyclopedias, and other books of the period show that it included trade: business in which persons bought and sold, bargained and contracted.  And this meaning has persisted to modern times. Surely, therefore, a heavy burden is on him who asserts that the plenary power which the Commerce Clause grants to Congress to regulate "Commerce among the several States" does not include the power to regulate trading in insurance to the same extent that it includes power to regulate other trades or businesses conducted across state lines.  
Justice Black argued that one of several reasons causing the courts to uphold the precedent set by PV was to ensure that at least some one, anyone, was regulating the industry,
In all cases in which the Court has relied upon the proposition that "the business of insurance is not commerce," it [sic] attention was focused on the validity of state statutes the extent to which the Commerce Clause automatically deprived states of the power to regulate the insurance business. Since Congress had at no time attempted to control the insurance business, invalidation of the state statutes would practically have been equivalent to granting insurance companies engaged in interstate activities a blanket license to operate without legal restraint. As early as 1866, the insurance trade, though still in its infancy, was subject to widespread abuses. To meet the imperative need for correction of these abuses, the various state legislatures, including that of Virginia, passed regulatory legislation.  Paul v. Virginia upheld one of Virginia's statutes. To uphold insurance laws of other states, including tax laws, Paul v. Virginia's generalization and reasoning have been consistently adhered to.
The practicality of the case was not just regulating and taxing authority but applicability of federal anti trust legislation to the insurance industry.  The two opposing schools of thought on the matter being: 1) Unregulated competition in the insurance industry (i.e. exemption from anti-trust) is not good for the public interest; and 2)  The business of insurance is different from all other forms of commerce and therefore the anti-trust laws ought not apply.

Justice Black concluded that the decision to regulate or not to regulate the insurance industry was not for the Court to decide, though he did state that the Court regarded the argument that the Sherman Act invalidated state laws as "exaggerated."  On the basis that insurance is commerce the Court reversed the PV ruling.  Congress was then free to regulate the insurance industry however they deemed fit.

The Supreme Court made their USSEA ruling on June 5, 1944- the day before D-Day.  As Congress began dealing with legislation to address their new found responsibility, the NAIC argued for continued regulation and taxation by the state and not the federal government and proposed a bill to Senators Pat Mccarran and Warren Ferguson which ultimately became the McCarran-Ferguson Act.

The MFA was accepted by Congress with little change.  Jones (2004) argues that MFA was so readily accepted because southern states were resentful towards New Deal policies and the stirring of civil rights debate brought up by WWII.  This gave them reason to fight for the preservation of state regulating authority based on feelings about state rights.  The north and midwest had an interest in preserving state regulating authority because most of the insurance companies were headquartered in north and midwest states and they were enjoying the tax revenue.  

The resulting MFA preserved state regulating and taxing authority of the states.  Federal antitrust regulation applies to insurance "to the extent that such business is not regulated by the state."  I have seen it argued that the MFA works to partially exempt insurers from anti trust legislation because it is unclear as to how best to interpret the above clause.

The NAIC then began drafting model laws for use by the states. 

As Randall (1998) explains the NAIC has a conflicting duel role of centralizing regulation and preserving state regulating authority.  The NAIC is not a government organization, has no accountability to voters, is not subject to government oversight, and is directly funded by the insurance industry.  The NAIC acts to centralize insurance regulation enough but not so much as to impose undesired federal legislation upon the industry.

So, why then?
It seems the reason states regulate insurance for two main reasons:
  1. Deep rooted american values regarding the division of power between state and federal government
  2. The need or desire to readily negotiate and manipulate localized market competition to meet the needs of the state and its economy at any given time
To a lesser degree, a third reasons is found in the political power and influence of the NAIC.

So, what then?
For one thing, the self interested power of the NAIC and their ability to effectively regulate ought to get further attention.  I may explore this in the future.

The main take away however, is that there is no historic precedent for a lack of public political will to share risk.  Indeed, as part of the Federalist Papers #45, James Madison argued,
The operations of the federal government will be most extensive and important in times of war and danger; those of the State governments, in times of peace and security.
States regulate insurance so that they can respond, uphold and encourage the public values and behavior for which they choose.  And presumably, they like having the tax revenue.  The set up also seems favorable for federal government because it waters down the power the insurance industry would otherwise amass if it could effectively organize at a national level.  

Though I have yet to look into the Congressional debate over the different bills for national and regional catastrophe coverage pools, I now have a historical context for understanding whatever I find.  If the proposals get bound up in Civil War era dialog of state rights then the status quo holds.  However, if the dialog has changed then it is worth understanding how and why it changed after remaining the same for over 100 years.  

Tuesday, February 18, 2014

The relative cost of S. Florida Homeowners Insurance: 1975 vs. 2013

While I was working on my dissertation I made a visit to the main branch of the Miami-Dade Public Library and dug through old newspaper articles about property insurance in Florida.  I was looking for historical context for current political debate about the state's homeowners insurance troubles. 

The image above is from a Miami Herald article by Molly Sinclair dated July 20, 1975.  It offers a starting point for evaluating how much homeowners insurance has increased over the years. 

Comparing yesteryears costs of insurance to today's costs is difficult because so much has changed: the cost of the home insured, the density and wealth of a place, how geophysical hurricane risk is measured, market conditions, etc.  

Still, we can draw lessons about the changing cost of insurance and gain understanding of context for political debate about the cost of insurance by looking at changes in cost relative to some relevant social dimension.  Say, for instance median income.    


The graph above shows the average cost of homeowners insurance from Dade, Broward, and Palm Beach Counties as a proportion of median income in 1975 and 2013.  I used the Miami Herald article for data on 1975 costs and the FLOIR's CHOICES website for data on 2013 costs (customer service at FLOIR assured me that the CHOICES website reflects second half or 4th quarter 2013 costs).  I used the US Census for median income data.  Also, the Miami Herald article cited costs for a $35k home.  Adjusted to 2013 values using the HPI for relevant metropolitan areas, this gives a home value range of ~$150-$200k.  So, I used the data on the Choices website for pre-2001 construction which have a value of $150k.  

The graph shows that the cost of homeowners insurance today is not just more than in 1975 but as a proportion of income, it is a substantially heftier expenditure for the household.  In 1975, homeowners insurance accounted for 1% of household median income whereas today it accounts for between ~8-10% of median income.

One take away from this is that the pace of increasing costs of homeowners insurance in Florida surpasses the rate of income growth.  An underlying issue in political struggles over the cost of insurance is not just how much the insurance costs in real terms but the burden of the cost relative to incomes. 

Wednesday, January 22, 2014

"The Collins Report"


In 1995, the Florida legislature created the Academic Task Force on Hurricane Catastrophe Insurance and directed the South Florida think-tank, Collins Center for Public Policy, to assist in the endeavor.

The legislature created the Task Force with the following directive (as reported in the final report):
The purpose of the Task Force shall be to comprehensively review the system of Florida residential property insurance markets, the Florida Hurricane Catastrophe Fund, the Florida voluntary residential property insurance market and the international reinsurance market, and make recommendations as to how said system can be best coordinated and/or restructured to meet the following needs: 
    1. To ensure that adequate hurricane catastrophe insurance coverage is available to Florida residents at an affordable price; 
    2. To ensure that in the event of a hurricane catastrophe adequate public and/or private sector insurance and reinsurance is available to pay the claims of Florida residents;
    3. To provide necessary incentives for private insurers and reinsurers to increase residential property underwriting in Florida; and
    4. To ensure that the Florida system of residual residential insurance markets and the Florida Hurricane Catastrophe fund do not act as disincentives for insurers and reinsurers seeking to underwrite residential property insurance.
The result of the $500k ($ 1995) appropriation was the "Final Report by the Academic Task Force on Hurricane Catastrophe Insurance: Restoring Florida's Paradise," or more commonly, the "Collins Report."

The Report outlined 15 policy options intended to work together as components of "A Balanced Equation for Florida's Future" which in turn were intended to produce a "healthy, competitive private insurance market."

The Report then established criteria for identifying a "healthy, competitive private insurance market":
  1. Coverage by financially strong private companies of most of Florida's homeowners for hurricane risks;   
  2. Affordable, competitive rates consisted with widespread coverage of homeowners by private companies;  
  3. Low numbers of the "truly uninsurable" in a single remaining Windstorm JUA; 
  4. A strong Hurricane Catastrophe Fund; and 
  5. Reasonable maximum market shares for any one company in high- risk regions of the state.
Now, there is much that can be said about the way the report turned out.  For instance, a) I'm not entirely convinced that the Task Force responded to legislative objectives; b) The idea of a truly uninsurable is a matter of much opinion and the criterion offers little direction; and c) Given the abundance of economic literature on competitive markets why did the Task Force create their own criteria???

But these are not the points I wish to take up here.

For the breadth of issues and analysis covered by the Collins Report I find it remarkable that the Task Force did not cover nor touch on the pivotal issue of the time- the introduction of catastrophe models for insurance pricing.

The models introduced dramatic changes to the flow, acquisition and availability of information about hurricane risk in the insurance market.  So significant was the issue that in the same year that the Florida legislature created the appropriation for the Task Force it also created the FCHLPM to regulate how, what and when information about the hurricane risk could be used in ratemaking.

Nobel Laureate, Joseph Stiglitz, known for his work on information asymmetry, argued that it is quite common for issues surrounding information to lead to absent or imperfect markets.  Though information flow may result in market power for some it may also come at the expense of social well being.  In his award acceptance speech Stiglitz explained,
One of the arguments for unfettered capital markets was that there were strong incentives to gather information; if one discovered that some stock was more valuable than others thought, if you bought it before they discovered the information, then you would make a capital gain. This price discovery function of capital markets was often advertised as one of its strengths. But the issue was, while the individual who discovered the information a nano-second before any one else might be better off, was society as a whole better off: if having the information a nano-second earlier did not lead to a change in real decisions (e.g. concerning investment), then it was largely redistributive, with the gains of those obtaining the information occurring at the expense of others. 
He continued to explain that this presented an opportunity for the state to intervene and protect the public,
There are potentially other inefficiencies associated with information acquisition. Information can have adverse effects on volatility. And information can lead to the destruction of markets, in ways which lead to adverse effects on welfare. We described earlier how the existence of asymmetries of information can destroy markets. Individuals sometimes have incentives to obtain information (creating an asymmetry of information), which then leads to the destruction of insurance markets, and an overall lowering of welfare. Welfare might be increased if the acquisition of this kind of information could be proscribed. Recently, such issues have become sources of real policy concern, in the arena of genetic testing. Even when information is available, there are issues concerning its use, with the use of certain kinds of information having either a discriminatory intent or effect, in circumstances in which such direct discrimination itself would be prohibited.
By proposing that Florida could have a "healthy, competitive insurance market" without acknowledging that struggles with new information and technology undermined competitive market ideals, the Task Force established a false pretense.

But, what does it matter what is written in a 20 year old report by a now defunct think-tank?

The Florida Catastrophic Storm Risk Management Center at FSU was created by the Florida legislature in 2007 to produce research that amongst other things, "are expected to have an immediate impact on policy and practices related to catastrophic storm preparedness."

The Center uses the Collins Report as a basis for its continued existence and for judging the state of Florida's property insurance market.  Because of the Collins Report shortcomings in addressing the information challenges facing Florida's insurance market it is no wonder that the Center routinely finds Florida as failing to meet the Task Force's criteria of a healthy competitive market.

Furthermore, using the Collins Report in this way narrows the scope of research activity, improved understanding and potential solutions to the challenges facing Florida and its management of catastrophic hurricane risk addressed by the Storm Risk Center.

Overlooking the effect information has on Florida's insurance market leads to research outcomes that consistently stack the deck in favor of those that would rather not go into detail about how the models are used, misused and the extent of uncertainty involved in their production and output.

Omitting of consideration of the role of information in Florida's insurance market failures also overlooks the scope of issues that legislatures were hoping to address with the creation and management of Citizens.

For example, with the roll-out of near term models onto the market in 2006 information asymmetry worsened in Florida because knowledge about the risk became controversial (more on this perhaps some other time).

The following year, during deliberation of the controversial HB1A (which placed Citizens into a competitive position in the marker), Rep. Denise Grimsley, a co-sponsor of the bill, argued that the proposed changes responded to a “competitive disadvantage” policyholders’ had when dealing with their insurance providers due to information asymmetry and industry folly:
Policyholders have too few options, too few protections, and too little information. Today, policyholders no longer stand on a level playing field with their insurers. The purpose of this legislation is to restore balance and common sense to the market
(Hearing recording, House Policy and Budget Council, FL HR, January 17, 2007).
For all the insights and ideas that the Collins Report brought to light it really failed to acknowledge a very important issue facing the Florida insurance market.  Yet, the report continues to have an impact by structuring how researchers evaluate and understand Florida's insurance market.

Ultimately, this restricts the tools available to policy makers to improve conditions.

Tuesday, December 17, 2013

Florida's Hurricane Risk is Overdetermined



The above images shows Wordles or word clouds of the first authors listed in the Meteorological references section of individual modeling groups submissions to the FCHLPM under 2011 standards.  You can find the same sort of word clouds for 2009 standards here.

Unique mixtures of science result in different hypothetical storms resulting in different hypothetical interactions with society.  


The above graph shows the differences in the radius of maximum winds between modeling groups.  The radius of maximum winds (denoted as Rmax) is a measure of the distance from the center of a hurricane to its maximum winds.  The mathematical derivation of Rmax depends upon central pressure.  Central pressure is calculated differently by company and judged reasonable by  respective company... often by the same statistical metrics.  

The graph depicts the range between a company's minimum Rmax and maximum Rmax for a given central pressure (CP). The difference is indexed to the average of the ranges for all companies for a given central pressure.  So, for example, at a central pressure of 920 mb, storms created by EQECAT have an Rmax anywhere from 4 miles to 62 miles (see data below), a range of 58 miles.  The average range for all the models at 920 mb is 35.42.  So, EQECAT's Rmax range is 60% greater than the average range at 920mb of all companies.  
Rmax values submitted to FCHLPM (All distances in Miles)
No matter.  The FCHLPM considers all of these accurate and reliable.  So too, the output.  


The table above shows the uncertainty interval of each company's 100 PML and 250 PML.  Given the exposure data, EQECAT estimates a 1% annual chance of a loss of within the range of $35B and $104B.  They have a 95% statistical confidence about this.  

Yet under 2009 standards and the same data set they had 95% confidence that the loss would fall between $21B and $122B. You can see 2009 standards data in the below table (I make no adjustments for inflation).  You can see this type of variation for any of the companies- I'm just using EQECAT as an example.


All of the data here is considered accurate, reliable and my favorite "reasonable."  Consider further that outside of Florida, insurers and others use totally different numbers and catalogs that they find to be accurate, reliable and reasonable. 

So my point here is that the risk is overdetermined.  This means that there are more scientific reasons and acceptable estimates than observations needed for determining the scientific quality of risk estimates.  

Given the substantial variation in accurate estimates of risk, the right or best estimate is determined by external factors such as, market and political conditions. Determining hurricane risk is, as this analysis shows, well suited for negotiation.