CIPR Newsletter June 2016 › cipr_newsletter_archive › vol19.pdfData at a Glance: U.S. Insured Catastrophe Losses 16 The severity of catastrophic events in 2011 and 2012 caused

July 2016 | CIPR Newsle er

JULY 2016

Eric Nordman CIPR Director 816-783-8232

[email protected]

Kris DeFrain Director, Research & Actuarial

816-783-8229 [email protected]

Shanique (Nikki) Hall Manager, CIPR 212-386-1930 [email protected]

Dimitris Karapiperis Research Analyst III


Anne Obersteadt Senior Researcher


NAIC Central Office Center for Insurance Policy and Research 1100 Walnut Street, Suite 1500 Kansas City, MO 64106-2197 Phone: 816-842-3600 Fax: 816-783-8175 h p://cipr.naic.org

Inside this Issue

U.S. Insurance Regula on: Group Supervision 2 As part of the Solvency Moderniza on Ini a ve, the NAIC Group Solvency Issues (EX) Work-ing Group expanded the “windows and walls” approach to improve regula on of U.S.-based insurers opera ng within corporate groups. Since 2010, many improvements to the U.S. group supervisory framework have been adopted by the NAIC and implemented in the states. This ar cle describes the group supervisory structure and some ongoing discussions related to group supervision. Why My Children Can S ll Watch the Peanuts Movie 6 FSOC’s Designa on of MetLife Disregarded the State Insurance Regulatory System The myth that federal regulators know best has long been dispelled, but it is s ll front and center today, par cularly when those same federal officials are challenged on their deci-sions. Alarm bells are going off at the Treasury Department a er a U.S. District Court recent-ly threw out the designa on of MetLife as a Systemically Important Financial Ins tu on (SIFI). This ar cle addresses the designa on of MetLife as a SIFI by the Financial Stability Oversight Council (FSOC). The NAIC Securi es Valua on Office 11 One important component of the na onal system of state-based insurance regula on is careful monitoring of the investments insurers hold, which enables them to make good on the promises they make to their policyholders. The purpose of this ar cle is to explore one of the many tools insurance regulators use to assist them with this important func on. It focus-es on the Securi es Valua on Office, which is responsible for the day-to-day credit quality assessment and valua on of debt securi es owned by state-regulated insurers. The Workers’ Compensa on Opt-Out Debate Con nues 12 Oklahoma Senate Bill 1062, also known as the Oklahoma Op on, allows “qualified employ-ers” to opt-out of the tradi onal workers’ compensa on system. The Oklahoma Op on rep-resents a significant development for the U.S. workers’ compensa on systems, which is com-pulsory for the majority of employers in all states except Texas. This ar cle provides an over-view of the Oklahoma Op on and also summarizes two recent studies on the opt-out debate. Data at a Glance: U.S. Insured Catastrophe Losses 16 The severity of catastrophic events in 2011 and 2012 caused an increase in the amount of insured losses, as well as an increase in homeowners direct loss ra os. Conversely, the years 2013–2015 were less than average for severe catastrophic losses. This ar cle features data for direct loss ra os for the homeowners’ lines of business and discusses the impact of catas-trophes on those loss ra os. Insurance: The Highest and Best Use Case for Blockchain Technology 17 Blockchain technology carries a promise of great opportunity and efficiency in business op-era ons and governance. This ar cle will introduce the reader to blockchain technology and explore how it may be used to assist with iden fying and facilita ng risk transfers in the insurance industry. In addi on, this ar cle demonstrates how blockchain technology could be integrated broadly and uniformly across society as well as the implica ons for the insur-ance industry.

2 July 2016 | CIPR Newsle er

U.S. I R : G S

By Kris DeFrain, Director, Research and Actuarial Department I U.S. insurance group supervision is o en described as a “windows and walls” approach, with “windows” referring to the ability of regulators to evaluate group ac vity and as-sess its poten al impact on the individual insurer and “walls” referring to the required regulatory approval of cer-tain monetary transac ons which help protect the capital of the insurer. As part of the Solvency Moderniza on Ini a ve, the NAIC Group Solvency Issues (EX) Working Group expanded the “windows and walls” approach to improve regula on of U.S.-based insurers opera ng within corporate groups. Since 2010, many improvements to the U.S. group supervi-sory framework have been adopted by the NAIC and imple-mented in the states. Among the major changes are: • Improving communica on between regulators; • Incorpora ng supervisory colleges into the regulatory

review processes; • Gaining access and collec ng necessary informa on

regarding en es in the group; • Clarifying and incorpora ng enforcement measures; • Reviewing and assessing capital on a group basis; and

expanding the accredita on standard.1 This ar cle describes the group supervisory structure and some ongoing discussions related to group supervision. U.S. G S F Group-wide insurance supervision in the U.S. is a risk-focused approach, addressing specific risks through legisla-

on or regula on and incorpora ng a supervisory review process to address the broad-based or unique risks. “In the U.S., group supervision and oversight is conducted primarily through licensed en es during the quarterly financial anal-ysis monitoring efforts resul ng from the implementa on and execu on of the Insurance Holding Company System Model Laws and Regula ons (IHC).”2 Group-wide regula on and supervision includes all insurers; all opera ng and non-opera ng holding companies; non-regulated en es; other regulated en es such as banks, u li es, and securi es companies; and special-purpose en es. The state-based system of group regula on is framed in the Model Law on Examina ons (#390), the Insurance Holding Company System Regulatory Act (#440), the Insurance Hold-ing Company System Model Regula on with Repor ng Forms and Instruc ons (#450) and the Risk Management and Own Risk and Solvency Assessment Model Act (#505).

Model #440 is the founda onal component and lays out the power for state insurance commissioner to regulate any registered insurer and its affiliates to assist in monitoring the con nued financial solvency of the insurer. All states and the District of Columbia have adopted substan ally similar lan-guage found within Model #440 and its related regula on Model #450, as required by the NAIC Financial Regula on Standards and Accredita on Program. Almost all states have adopted substan ally similar language to Model #505, with approximately 90% of U.S. premium subject to repor ng an annual Own Risk and Solvency Assessment (ORSA) Summary Report. All states are expected to have this adopted before it becomes an accredita on requirement in 2018. Supervision of the holding company system is applied using mul ple mechanisms: 1) repor ng requirements; 2) licensing oversight; 3) financial analysis; and 4) financial examina on review procedures. Guidance to insurance groups and state insurance regulators are provided in numerous NAIC tools, including: 1) financial statement repor ng requirements; 2) ORSA requirements and guidance; 3) financial analysis proce-dures; and 4) financial examina on procedures. The NAIC Financial Analysis (E) Working Group provides an addi onal layer of surveillance for insurance groups overall, supplemen ng individual state insurance departments’ sol-vency monitoring. The Working Group performs quarterly analysis on na onally significant groups that exhibit charac-teris cs of trending toward or being financially troubled. The Working Group then works with domiciliary regulators and the lead state to advise the most appropriate regulato-ry strategies, methods and ac ons. This system provides peer review on surveillance of insurance groups to ensure quality analysis. M G S S 2010: “W W ” In 2010, following the global financial crisis, the NAIC estab-lished a plan of ac on to address evolu onary developments in the insurance industry and improve group supervision. The plan was established to build on the strengths of the U.S. legal-en ty supervisory approach and take the following ac ons: 1) improve communica on between regulators; 2) incorporate supervisory colleges into regulatory review pro-cesses; 3) gain access to and collect needed informa on regarding en es in the group; 4) clarify and incorporate

(Continued on page 3) 1 Group Solvency Issues (EX) Working Group, Feb. 26, 2010, “Report to Solvency

Moderniza on Ini a ve (EX) Task Force on Suggested ‘Windows and Walls’ Ap-proach for Regula on of United States Based Insurers Opera ng within Corporate Groups”

2 Ibid.

July 2016 | CIPR Newsle er 3

U.S. I R : G S (C )

enforcement measures; 5) review and assess capital on a group basis; and 6) expand the accredita on standard. Communica on between regulators. Iden fied in 2010 as the first and most important component of group supervi-sion of regulated en es, significant a en on was placed on improving communica on with the primary group regu-lator. Lead states have been established for each U.S.-based group, and those states have entered into informa on-sharing agreements and memoranda of understanding (MOUs) with interna onal regulators/supervisors. The lead state approach for insurance groups and associat-ed best prac ces are documented in the Financial Analysis Handbook. The lead state is selected by the domes c state insurance regulators of the group, generally by consensus. The lead state is responsible for establishing communica-

on with other iden fied states, federal regulators and in-terna onal regulators, and it must perform numerous ex-amina on procedures. Lead states must analyze the finan-cial condi on of the insurance holding company system in four different areas: 1) profitability; 2) leverage; 3) liquidity; and 4) the overall financial condi on. Lead states must determine and document: 1) the depth of the insurance holding company analysis; 2) the assessment of the group’s governance and enterprise risk; 3) ques ons addressed in periodic mee ngs with the group; 4) target examina on procedures; and 5) the extent to which there are any market conduct risks. Most importantly, the lead state works with other regulators to determine what, if any, further ac on is appropriate regarding the domes c insur-ers in the group or the group as a whole. The lead state will prepare a Group Profile Summary, providing a high-level overview of: 1) the group’s financial condi on; 2) the group’s strengths and weaknesses; 3) the group’s current and prospec ve risks; and 4) the ongoing regulatory plan to ensure effec ve supervision. State insurance regulators will provide no fica on about troubled insurers to each other and to federal and interna-

onal regulators when the insurer is opera ng in a group with en es subject to federal or interna onal oversight. It is safe to say improved communica on between regula-tors will remain a goal of insurance regulators everywhere, and there is s ll room for improvement to enhance coordi-na on and coopera on with federal and interna onal regu-lators and supervisors. Supervisory colleges. “Supervisory colleges are joint mee ngs of interested regulators with company officials

and include detailed discussions about financial data, corpo-rate governance and enterprise risk management (ERM) func ons. Supervisory colleges are intended to facilitate oversight of interna onally ac ve insurance companies at the group level. The Interna onal Associa on of Insurance Supervisors (IAIS) defines a supervisory college as ‘a forum for coopera on and communica on between the involved supervisors established for the fundamental purpose of fa-cilita ng the effec veness of supervision of en es which belong to an insurance group; facilita ng both the supervi-sion of the group as a whole on a group-wide basis and im-proving the legal en ty supervision of the en es within the insurance group.’”3 The aim of supervisory colleges is to share informa on gath-ered by all regulators to improve the oversight of group risks, as well as provide clear channels of communica on to be used to navigate through any poten al solvency con-cerns. U.S. state insurance regulators believe supervisory colleges achieve this aim and have proven to be useful pla orms to improve supervisory coopera on and coordina-

on between interna onal regulators. Supervisory colleges have been formally incorporated into regular review processes of interna onally ac ve groups via IHC enhancement and regulator best prac ces documented in the Financial Analysis Handbook. U.S. lead states, when designated as the group-wide supervisor, are ac vely con-vening supervisory colleges for all groups iden fied as inter-na onally ac ve insurance groups (IAIGs). The U.S. lead state takes on the regulatory coordina on role for a U.S. interna onally ac ve insurer and assists with development of “crisis management plans” to support the management of any arising crisis situa on. Access to and collec on of informa on. State laws require annual filings regarding the holding company system detail-ing intercompany contract terms, rela onships, biographical and other data for officers and directors of the ul mate par-ent, and other financial informa on. Addi onal holding company financial informa on is required through other statutory filings such as the NAIC financial annual state-ment, which includes holding company informa on such as disclosure of affiliated transac ons and a detailed organiza-

onal chart (Schedule Y). The Insurance Holding Company System Regulatory Act (#440) provides access to books and records of the holding company system and affiliates to as-certain the financial condi on of the insurer.

(Continued on page 4)

3 CIPR website: www.naic.org/cipr_topics/topic_supervisory_college.htm.


U.S. I R : G S (C )

Group supervision was strengthened with revisions adopted in the Insurance Holding Company System Regulatory Act (#440) and the Insurance Holding Company System Model Regula on with Repor ng Forms and Instruc ons (#450) in 2010. The revisions included the following: 1) expanded ability to evaluate any en ty within an insurance holding company system; 2) enhancements to the regulator’s rights to access books and records and compelling produc on of informa on; 3) establishment of funding with regard to regulator par cipa on in supervisory colleges; and 4) en-hancements in corporate governance, such as board of di-rectors and senior management responsibili es. State insur-ance regulators also adopted an expansion to the Insurance Holding Company System Annual Registra on Statement (Form B) to broaden repor ng requirements to include fi-nancial statements of all affiliates. A new Form F (Enterprise Risk Report) was required for firms to iden fy and report their enterprise risk. In the financial analysis process, state insurance regulators will perform rou ne analysis on holding companies. Infor-ma on collected during the process includes the following: 1) Form A—Statement of Acquisi on of Control of or Mer-ger with a Domes c Insurer; 2) Form B—Insurance Holding Company System Annual Registra on Statement including financial statements; 3) Form D—Prior No ce of a Transac-

on (for regulatory approval of management agreements, service contracts, tax alloca on agreements, guarantees, loans and all cost-sharing arrangements, and approval of risk transfer in reinsurance agreements); 4) Form E (or Oth-er Required Informa on)—Pre-Acquisi on No fica on Form Regarding the Poten al Compe ve Impact of a Pro-posed Merger or Acquisi on by a Non-Domiciliary Insurer Doing Business in This State or by a Domes c Insurer; 5) Form F—Enterprise Risk Report; 6) extraordinary dividend distribu on; and 7) ORSA Summary Report. Addi onal infor-ma on can be gathered by the lead state as necessary to evaluate the group. “The Form F requires the ul mate controlling person to iden-

fy the material risks within the insurance holding company system that could pose enterprise risk to the insurer.”4 The ORSA requires repor ng on reasonably foreseeable and rele-vant material risks to which an insurer or insurers are ex-posed. State insurance regulators are currently discussing an overlap of repor ng requirements for the Form F and ORSA. While the ORSA focus is on risks that insurers are exposed to and Form F focus is on enterprise risks to the en re group, some repor ng overlap exists since risks that directly affect the insurer may also qualify as enterprise risks to the whole group. State insurance regulators will be considering the differences and similari es in the scope of repor ng require-

ments and whether enterprise risks that do not directly affect an individual insurer are being reported in ORSA filings. State insurance regulators will also consider the extent of overlap given the ORSA exemp on for smaller insurers/groups. Enforcement measures. The aim of regulatory enforcement is to protect the insurer and its policyholders when viola-

ons occur. In 2010, a legal review of the exis ng authority generally showed li le need to seek addi onal enforcement authority. However, there were two main changes made to enforcement authority. Revisions to the Insurance Holding Company System Regula-tory Act (#440) strengthened enforcement with the addi on of more authority to order or compel informa on to be pro-vided about the Holding company. Viola ons already includ-ed failure to file required material related to a empts to acquire control of, or merge with, a domes c insurer with-out commissioner approval, but the first change expanded these to include any a empts regarding “dives ture of” a domes c insurer without commissioner approval. The sec-ond change expanded the sanc ons in response to any vio-la on of Sec on 3 (Acquisi on of Control of or Merger with Domes c Insurer) of Model #440 that prevents the regula-tor from having a full understanding of the enterprise risk to the insurer by affiliates or by the insurance holding company system. Any such viola on now serves as an independent basis for the regulator to disapprove dividends or distribu-

ons from a legal en ty insurer. Group capital assessment. In 2010, the NAIC expanded the analysis procedures domiciliary regulators u lize to monitor legal en ty insurers with a set of procedures specific to group analysis performed by the lead state of the insurance group. A key element of the group analysis procedures is assessing the adequacy of group capital. Group capital assessment in-forms the lead state regulator not only of the current finan-cial posi on of the group, but it also provides a basis for the regulator to consider the impact of various proposed transac-

ons and poten al events. The regulator’s assessment is pri-marily generated from the significant amount of informa on and analy cal tool results for legal en ty insurers, the infor-ma on obtained from the Holding Company filings, and the results of the remaining group analysis procedures. To enhance the lead state regulator’s group capital assess-ment, most states have now implemented the ORSA require-ment providing insurance regulators a window into the insur-ance group’s view of group capital needs. While the ORSA calcula on of group capital is a quan ta ve calcula on, the

(Continued on page 5) 4 NAIC Financial Analysis Handbook—2015 Annual/2016 Quarterly, Sec on V.F.


U.S. I R : G S (C )

regulatory review is a qualita ve approach. The lead state will “develop and document the general methodology applied and how outputs from the prospec ve solvency calcula ons compare with recent trends for the group and, in general, be able to determine the sufficiency of capital.”5

With the qualita ve group capital approach implemented by the states, the NAIC is now exploring poten al group capital quan ta ve approaches. In late 2015, the NAIC decided to embark on development of an aggregated RBC methodology as a calcula on to assess group capital and to do so u lizing the group capital assessment approach rather than crea ng a minimum group capital requirement or standard. The NAIC Group Capital Calcula on (E) Working Group is developing a consistent methodology for state insurance regulators to u lize when assessing a U.S. insurance group’s capital as part of their lead state responsibili es. The group capital calcula on would then be able to serve as a baseline quan ta ve measure to be used by regulators in conjunc-

on with assessment of group-specific risks and stresses that may not be captured in legal en ty RBC filings. Current plans are to use a legal-en ty RBC aggrega on approach rather than develop addi onal measures. There are many outstanding decisions to be made, including how to address legal en es within the group that are not currently subject to minimum capital requirements. The Working Group aims to see significant development on the group capital calcula-

on methodology by year end. Accredita on. “Accredita on is a cer fica on given to a state insurance department once it has demonstrated it has met and con nues to meet an assortment of legal, financial and organiza onal standards as determined by a commi ee of its peers.”6 More detailed annual holding company analysis was required of all states as an accredita on standard (including many of the steps outlined under “Communica on Between Regula-tors” on page 3). The 2010 revisions to the Insurance Holding Company System Regulatory Act (#440) (which clarified ac-cess to group informa on, provided for supervisory colleges, required Form F – Enterprise Risk Report, etc.) were required for accredita on in 2016. The 2014 Model #440 revisions (allowing a group-wide supervisor role with corresponding authority and responsibili es) are currently under considera-

on as an addi onal accredita on requirement, with a likely effec ve date of 2020. And the Risk Management and Own Risk and Solvency Assessment Model Act (#505) requiring ORSA will become an accredita on standard in 2018.

C By law, obliga ons to policyholders con nue to be made and enforced at the legal en ty level. Thus, within a group se ng, U.S. state insurance regulators’ core focus is on le-gal en es. However, with group holding company level or affiliate ac ons poten ally affec ng the insurance compa-ny’s ability to pay claims, U.S. state insurance regulators implemented a “windows and walls” approach for the regu-la on of U.S.-based insurers opera ng within corporate groups. Significant changes have been implemented since 2010 to enlarge the windows and strengthen the walls, in-cluding changes made to require repor ng of enterprise risk, assessment of group capital and implementa on of supervisory colleges. State insurance regulators will con nue to assess and im-prove on the group supervisory framework over me. One major work stream s ll underway at the NAIC is the imple-menta on of a uniform approach to quan ta ve calcula on for group capital assessment. Addi onal changes will be considered in light of the depth of the overall regulatory framework in the U.S.; the legal framework for regulatory ac on; the protec on of policyholders at the en ty level; and the absence of a clear path to the flow (fungibility) of capital between en es regulated by different jurisdic ons and opera ng under different laws.

A A

Kris DeFrain is the NAIC Director of the Re-search and Actuarial Department. She is cur-rently charged as primary NAIC staff for the Principle-Based Reserving and the Casualty Actuarial and Sta s cal Task Forces.

Ms. DeFrain manages a staff of actuaries, sta s cal analysts, insur-ance contract experts, and research analysts working on regulatory solvency- and market-related issues, providing regulatory services, and conduc ng research for the Center for Insurance Policy and Research. Ms. DeFrain received her bachelor’s degree in finance/actuarial science from the University of Nebraska in 1989. She received her FCAS designa on from the Casualty Actuarial Society (CAS), where she previously served as Vice President—Interna onal. Ms. DeFrain is a member of the American Academy of Actuaries and a Char-tered Property & Casualty Underwriter.

5 NAIC Financial Analysis Handbook—2015 Annual/2016 Quarterly, Sec on V.

6 CIPR website: www.naic.org/cipr_topics/topic_accredita on.htm.


W M C C S W P M FSOC' D M L D S I R S

By Nick Gerhart, Iowa Insurance Commissioner Originally published on May 16, 2016. Reprinted with per-mission from the Iowa Insurance Division. A er my ten-year old daughter caught me reading “Snoopy the Destroyer” in Paul Krugman’s April 11, 2016 New York Times piece, I had to reassure her that Snoopy was s ll safe to watch. Paul Krugman may scare li le girls with his tle, but consumers and the broader economy should not be scared due to the work of state insurance regulators. First, the myth that federal regulators know best would have Charlie Brown saying “Good grief.” As Iowa’s state insurance commissioner, I work closely with 55 other insur-ance commissioners, directors or superintendents through the NAIC to supervise the na on’s insurance market. I serve as Chair of the NAIC Life and Annuity (A) Commi ee, which sets na onal policy on ma ers that impact the life insur-ance and annuity industry. It is through these roles that I know our state-based regulatory system works to protect all American consumers and the financial soundness of the carriers we regulate. The myth that federal regulators know best has long been dispelled, but it is s ll front and center today, par cularly when those same federal officials are challenged on their decisions. Loud alarm bells are going off at the Treasury Department a er a U.S. District Court recently threw out the designa on of MetLife as a Systemically Important Fi-nancial Ins tu on (SIFI). In the wake of the decision, Treasury Secretary Jack Lew lashed out at the court in public statements and in the pag-es of major U.S. newspapers that MetLife, a company that is almost 150 years old, somehow presents a threat to the U.S. unless the company is subjected to addi onal oversight and federal regula on that accompanies the SIFI designa-

on. A er following the coverage, it was abundantly clear that Secretary Lew fails to acknowledge that MetLife, Pru-den al and hundreds of other insurance companies are closely and successfully regulated and supervised by state insurance commissioners for decades. Cri cs of the court’s decision—including Lew—argue that the court did not defer to the “experts” at the Financial Stability Oversight Council (FSOC), which found that MetLife posed a threat to the economy. However, the FSOC itself ignored its own true experts. The FSOC designated MetLife as a SIFI in December of 2014 over the objec ons of the

only two independent FSOC members with true exper se and experience in insurance regula on. What we have learned post 2008 financial crisis—which Congress should keep in mind—is that the near collapse of AIG, and the accompanying $180 billion check from taxpay-ers to save it, was due to suspect federal oversight of AIG, not our state-based insurance regulatory system. AIG was a large complex financial company. Its financial products divi-sion was already supervised by the Office of Thri Supervi-sion (OTS). While the insurance opera ons were supervised by the state insurance regulators, it was AIG’s financial products division that became overextended by offering credit default swaps backed by the sizable balance sheet of AIG. OTS did not require AIG’s financial products division to have sufficient capital and the company was le exposed. I want to ensure that the economy is protected from a fi-nancial crisis as much as anyone, however, the thought of federal regula on and supervision of tradi onal insurance ac vi es does not allow me to sleep easier. To the contrary, we need a coordinated regulatory approach where regula-tors work together and compliment the work we all do. The federal government’s supervision shouldn’t expand into areas where it lacks exper se and where states are already working effec vely. Consumers will bear the costs of this unnecessary overlap and duplica on. The insurance industry is under the watchful eye of state insurance departments, and our track record of solvency regula on of large na onal and mul na onal insurers speaks for itself. The NAIC and state insurance departments have updated our solvency oversight through transforma-

onal model laws which allow state insurance regulators to maintain a close watch on the insurance ac vi es of the carriers domiciled in our respec ve states. When Judge Rosemary Collyer rescinded MetLife’s designa-

on, she found that FSOC failed to take into account exis ng regulatory regimes. In short, the court found that FSOC failed to consider the level of scru ny that insurance carri-ers like MetLife face from state regulators. Through strong statutory schemes and strong financial over-sight state regulators have detailed knowledge of insurance companies and successfully protect policyholders in all 50 states. Insurance companies do fail as part of a healthy and compe ve market, but it is a rare occurrence since state insurance regula ons require that insurance companies hold significant amounts of capital in reserve, and even



W M C C S W P M (C )

then, policyholders are priori zed and protected through our receivership process. In addi on to the amount of capi-tal a carrier is required to hold, states also maintain strict rules on the kinds of investments an insurance carrier may make with policyholders’ money. If the OTS maintained this same type of oversight and approach, AIG would not have required a massive taxpayer bailout. Notwithstanding the above, alarmists like Secretary Lew insist that a large insurance company poses a threat to the U.S. economy and thereby requires federal supervision. They claim that Judge Collyer’s decision throwing out Met-Life’s SIFI designa on undermines broader efforts to reform our financial system in the wake of the 2008 financial crisis. Judge Collyer’s decision is focused on the process that the Financial Stability Oversight Council used for designa ng MetLife as a SIFI in 2014, not the broader law. The federal Dodd-Frank Wall Street Reform and Consumer Protec on Act an cipated the fallibility of FSOC designa-

ons, as the law allows companies to challenge their desig-na on. I do not blame MetLife for challenging the FSOC in the manner it designated the company. FSOC con nues to dismiss the legacy and exper se of state regulators like myself and has announced that it will appeal the District Court’s decision. Given its record, it is puzzling why the federal government believes it can do a be er job than on-the-ground state commissioners. As state commis-sioners, our top priority is protec ng policyholders and providing robust oversight. Thankfully this allows me to tell my daughter that Snoopy isn’t the bad guy, but she may learn to fear alarmists like Krugman and Secretary Lew as she begins to read past the funny pages.

A A

Iowa Governor Terry Branstad appointed Nick Gerhart to serve as Insurance Commis-sioner beginning February 1, 2013. Gerhart serves on the NAIC Execu ve (EX) Com-mi ee, Life Insurance and Annui es (A) Com-mi ee, Financial Condi on (E) Commi ee and Interna onal Insurance Rela ons (G) Commi ee. In addi on, Gerhart is a board member of the NIPR.

Gerhart was named by the U.S. Department of the Treasury in 2014 to serve on the Federal Advisory Commi ee on Insurance (FACI), which advises the Federal Insurance Office (FIO) on domes c and interna onal policy. He has represented the NAIC at the Interna-

onal Associa on of Insurance Supervisors and at the EU-U.S. In-surance Dialogue Project. Gerhart is a recognized thought leader on health care ma ers and has been at the forefront during the implementa on of the federal Affordable Care Act (ACA). He also led the team that created and hosted the Global Insurance Symposium in Des Moines, IA, an event that a racted more than 350 people from all over the world to discuss emerging issues in insurance. Gerhart is passionate about educa ng Iowans on insurance and investment ma ers and creat-ed Iowa Fraud Fighters, a statewide ini a ve that aims to educate seniors on how to protect themselves and not fall prey to fraudu-lent investment scams.


T NAIC S V O

By Eric Nordman, Director of Regulatory Services and CIPR, Ramon Calderon, Interna onal Policy Director, and Jeff Johnston, Senior Director, Financial Regulatory Affairs—Domes c Policy and Implementa on I One important component of the na onal system of state-based insurance regula on is careful monitoring of the in-vestments insurers hold, which enables them to make good on the promises they make to their policyholders. The pur-pose of this ar cle is to explore one of the many tools insur-ance regulators use to assist them with this important func-

on. It focuses on the Securi es Valua on Office (SVO), one of three groups within the NAIC Capital Markets & Invest-ment Analysis Office. Located in New York City, the SVO is responsible for the day-to-day credit quality assessment and valua on of debt securi es owned by state-regulated insurers. Insurers are obligated to report ownership of securi es to the SVO when such securi es are required to be reported in the NAIC financial statements. All long-term bonds, pre-ferred stocks and common stocks held by insurers are re-quired to be reported in Schedule D. Separate reports of each are required for insurer holdings at the end of each year. Other schedules show long-term bonds and stocks acquired during the current repor ng year—all long-term bonds and stocks sold, redeemed or otherwise disposed of during the current year, and all long-term bonds and stocks acquired during the year and fully disposed of during the current year. All other long-term invested assets are report-ed in Schedule BA. All short-term investments are reported in Schedule DA. The SVO staff evaluates these securi es for the purpose of assigning NAIC-SVO credit quality designa ons (NAIC desig-na ons) and/or unit price. These designa ons and unit prices are produced solely for the benefit of state insur-ance regulators, who may use them as part of their moni-toring of the financial condi on of domiciliary insurers. A domiciliary insurer is an insurer who maintains its principle office in the state or jurisdic on (known as the domes c regulator). An insurer can be licensed in many jurisdic ons; but, it can only be a domiciliary insurer in one jurisdic on. C O T C R O There is a tendency for people to think of all credit ra ng organiza ons as performing the same basic func ons. There are several well-known private credit ra ng organiza ons called Na onally Recognized Sta s cal Ra ng Organiza ons

(NRSROs). The U.S. Securi es and Exchange Commission (SEC) is primarily responsible for enforcement of federal se-curi es laws and regula ng the securi es industry. Among the agency’s many du es is designa on of the NRSROs. An NRSRO is a credit ra ng agency that the SEC allows other financial firms to use for some regulatory purposes, such as measuring capital requirements or for insurers, ascertaining the strength of assets held in reserve for future claims. There are currently nine firms designated by the SEC as NRSROs. They are: • Standard & Poor’s (S&P) • Moody’s Investors Service • Fitch Ra ngs • Kroll Bond Ra ng Agency • A.M. Best Company • Dominion Bond Ra ng Service Ltd. • Japan Credit Ra ng Agency Ltd. • Egan-Jones Ra ng Company • Morningstar Inc. The NRSROs most visible in the insurance sector are A.M. Best, S&P, Moody’s and Fitch. It is important to note not all of the NRSROs provide ra ngs for U.S. insurers. Further, not all ra ng agencies providing ra ngs to insurers are designat-ed by the SEC as an NRSRO. An example of a non-NRSRO providing insurer financial strength ra ngs is Demotech Inc. Demotech provides ra ngs to property and casualty insur-ers, tle insurers, health insurers, risk reten on groups and self-funded en es. The purpose of a NRSRO designa on is for the convenience of the public, regulators and regulated en es. It helps banks, broker-dealers and insurers meet capital requirements. If a financial ins tu on is invested in highly liquid, safe securi es, less capital is needed to meet financial obliga ons. In the banking sector, it helps guard against a run on the bank. In the insurance sector, it is a ma er of solvency protec on. Unlike the ra ngs of NRSROs, NAIC designa ons are not produced to aid the investment decision-making process of the public or ins tu onal investors. The purpose of the NAIC designa on is to assist financial regulators with their evalua on of the level of confidence the regulator should have in the assets held by insurers. Coupled with laws re-quiring insurers to invest conserva vely, the NAIC designa-

ons allow the financial regulator to evaluate whether the insurer has sufficient, conserva vely invested assets to meet its obliga ons to policyholders. Therefore, NAIC des-



T NAIC S V O (C )

igna ons are not suitable for use by anyone other than insurance regulators. T NAIC D P The SVO staff provides technical advice to the NAIC Valua-

on of Securi es (E) Task Force, a subgroup of the Financial Condi on (E) Commi ee. The SVO also supports other NAIC task forces and working groups addressing insurer invested assets. The SVO is comprised of 35 investment professionals who assess, at least annually, the credit risk of certain unrated investments owned by insurers. The SVO staff assigns an NAIC designa on (1 through 6), a measure of credit risk, to investments required to be filed with the NAIC. In 2015, the SVO staff received securi es filings from 1,401 insurers. These filings covered 12,997 securi es with an insurer total carrying value of approximately $560 billion. The vast ma-jority of these filings consist of unrated bond debts issued by U.S. corpora ons, otherwise known as “true private placements.” The SVO staff also translates credit ra ngs from NRSROs to equivalent NAIC designa ons. Insurers owning assets rated by NRSROs are not required to file the assets with the NAIC for ra ng purposes. This capability allows financial regulators to assess invest-ment credit risks of unrated investments through their own independent, supervisor-driven process instead of relying exclusively on NRSROs. The NAIC designa ons and other analysis produced by the SVO staff are used in supervisory processes to monitor insurers, including the appropriate-ness of the RBC investment charges. Without an NAIC desig-na on, an insurer would be reluctant to invest in an un-rated security because of the poten al for the security to not be allowed as an admi ed asset. The SVO derives its authority from state investment laws and reports to the Valua on of Securi es (E) Task Force. C NAIC D NRSRO C R An NRSRO credit ra ng and an NAIC designa on are not iden cal. A credit ra ng is a communica on between an issuer, a credit ra ng agency and investors. The purpose of that communica on is to provide investors with the NRS-RO’s opinion as to the issuers and/or the issues’ credit risk. An issuer wan ng a credit ra ng hires an NRSRO to perform the analysis necessary to provide the credit ra ng to the issuer for a fee, with the understanding the issuer will use it to communicate with the market. The expecta on is inves-tors will use such a ra ng in their analysis of the issuer and

the par cular financial instrument the issuer is proposing to sell. Accordingly, the issuer effec vely needs to have a ra ng from a trusted source in order to sell the security to investors. An issuer cannot obtain an NAIC designa on from the SVO. Only U.S. insurers can submit a request for an NAIC designa-

on. To receive an NAIC designa on, an insurer has to first own the security—in other words, an investor has already made the decision to purchase the security. An NAIC desig-na on is a communica on between the SVO staff and the state insurance financial regulator. The purpose of the com-munica on is to convey an independent opinion of the qual-ity of the investment the insurer purchased. I R R U D All insurers owning the same security are required to use the same NAIC designa on for official repor ng in annual finan-cial statements to their state insurance regulator. The desig-na on is also used in the NAIC accoun ng valua on rules under the Statutory Statements of Principle (SSAP)—including, SSAP No. 26—Bonds, SSAP No. 30—Unaffiliated Common Stock and SSAP No. 32—Preferred Stock—and to determine the accoun ng prac ces to be followed for mod-eled and non-modeled residen al mortgage-backed securi-

es (RMBS) and commercial mortgage-backed securi es (CMBS) under SSAP No. 43R—Loan-Backed and Structured Securi es. NAIC designa ons are also used to establish RBC and investment categories under state investment laws, which limit the type and amount of admi ed assets that can be invested in certain designa on grades (e.g., non-investment grade). Insurers access the designa ons using the NAIC Automated Valua on Service (AVS+) system. The AVS+ is a Web-based applica on that allows insurers to maintain a por olio of securi es’ informa on with the NAIC. For each security in an insurer’s por olio, the applica on provides the NAIC desig-na on and review date, among other data. This por olio of informa on can then be downloaded and exported into so ware applica ons that facilitate the prepara on of the NAIC annual and quarterly statements filed with the NAIC and state insurance departments. The NAIC designa ons are maintained within proprietary NAIC systems. Insurers are allowed to access the system for the sole purpose of repor ng the designa ons to regulators. The only excep on is access provided to academics under a



T NAIC S V O (C )

license.1 The NAIC allows non-U.S. insurers access to AVS+ informa on through an individual look-up agreement.2 While general access to the AVS+ system is limited, an insur-er’s public statutory financial statements contain the desig-na on for the assets owned by the insurer. The NAIC estab-lished the access parameters of the NAIC designa ons to reflect the purpose for which they were created—assessing the credit risk of insurers’ investments as part of the regula-tory financial solvency monitoring framework. As discussed previously, while credit ra ngs and NAIC desig-na ons may seem similar, there are important dis nc ons. An investor that is not a U.S insurer who obtains and uses an NAIC designa on will find that it has differences compared to a credit ra ng because it was not produced for the same reason or for the same purpose. NAIC designa ons are mon-itored on a different basis than credit ra ngs. An NAIC desig-na on also reflects perspec ves unique to the objec ves of the NAIC and state insurance regulators, including: 1) wheth-er it is an admi ed asset; 2) whether it qualifies as an invest-ment eligible to receive a designa on under the SSAPs; 3) whether it contains addi onal non-payment risks; and 4) a variety of other nuances ed to insurers’ regulatory financial solvency and repor ng concerns. P I A NAIC D Some have asked whether NAIC designa ons could be used to sa sfy financial regulatory requirements in other coun-tries. Generally speaking, the U.S. issuer of the private placement would have to abide by the laws of the non-U.S. insurer’s home country. In other words, it must either regis-ter the offering or abide by the rules governing an exemp-

on from registra on in that country. If that same security were offered in either country, and done so in accordance with the laws of both, then it is possible that a non-U.S. insurer could own the same security as a U.S. insurer. However, as a prac cal ma er, costs and other complexi-

es make it unlikely that a U.S. private placement issuer has the incen ve to make the opportunity available to investors in other countries. In this regard, only a small amount (less than 5%) of the true private placement market is owned by non-U.S. insurers. The Interna onal Associa on of Insurance Supervisors (IAIS) is exploring the development of a risk-based global insur-ance capital standard (ICS). There have been discussions about whether the NAIC designa on could be used as a measurement tool for the ICS. To date, the use of NAIC des-igna ons for such purposes has never been considered by

A A Eric Nordman, CPCU, CIE, is the director of the NAIC Regulatory Services Division and the CIPR. He directs the Regulatory Services Division staff in a wide range of insurance research, financial and market regulatory ac vi es, suppor ng NAIC commi ees, task forces and working groups. He has been with the NAIC since 1991. Prior to his ap-pointment as director of the Regulatory

Services Division, Mr. Nordman was director of the Research Divi-sion and, before that, the NAIC senior regulatory specialist. Before joining the NAIC, he was with the Michigan Insurance Bureau for 13 years. Mr. Nordman earned a bachelor’s degree in mathema cs from Michigan State University. He is a member of the CPCU Socie-ty and the Insurance Regulatory Examiners Society.

Con nued on next page

the NAIC members. If the IAIS members wish to move for-ward with a proposal to use NAIC designa ons for ICS pur-poses, the appropriate NAIC commi ee would need to delib-erate on the ma er. Under current rules, the NAIC-SVO staff is precluded from performing credit analysis on an unrated security owned by a non-U.S. insurer. Conversely, we are unaware of any other jurisdic on or supervisor that authoriz-es the use of NAIC designa ons for any regulatory purpose. S The SVO of the NAIC has a long history of serving the needs of U.S. insurers and insurance regulators. State insurance regulators rely on the NAIC designa ons where there is no alterna ve means for assessing the credit quality of author-ized debt investments. The SVO designa on process and product are an integral part of the U.S. insurance regulatory system. If state insurance regulators, working through the NAIC, consider adop on of the IAIS ICS in the future, it is a near certainty that use of NAIC designa ons would con n-ue. This posi on is consistent with certain mandates of the federal Dodd-Frank Wall Street Reform and Consumer Pro-tec on Act, regarding reduced reliance on external ra ng agencies for regulatory purposes. For more informa on, please refer to the Purposes and Procedures Manual of the NAIC Investment Analysis Office (P&P Manual) and/or con-tact NAIC senior management.

1 Access provided to academics under a license includes the same restric ons im-posed on insurers and their third-party administrators (TPAs) and to insurer invest-ment managers.

2 This agreement provides an insurer with up-to 50 CUSIPs (assigned security num-bers) for a nominal per-CUSIP fee.


A A (C ’ ) Ramon Calderon joined the NAIC in August of 2009 to serve as Director. Prior to joining the NAIC, he served as Deputy Commissioner, Financial Surveillance Branch, for the Califor-nia Department of Insurance. As a regulator, Mr. Calderon chaired the NAIC’s Interna onal Solvency and Accoun ng Working Group, among other NAIC working groups. In his 30+

years of service with the California Department of Insurance, Mr. Calderon developed exper se in solvency regula on with an empha-sis on insurance accoun ng and business restructurings. He advised the California Insurance Commissioner on a wide range of financial surveillance issues, and served in the capacity of Deputy Commis-sioner un l his departure from the California Department of Insur-ance in August of 2009. In his current role at the NAIC, Mr. Calderon is ac ve in interna onal ma ers and works closely with state insurance regulators and the ComFrame Development and Analysis (G) Working Group (CDAWG) members, and is ac ve with the Interna onal Associa on of Insur-ance Supervisors (IAIS). He is a member of the IAIS Capital, Solvency and Field Tes ng Working Group and the Insurance Groups Working Group.

Jeff Johnston oversees all aspects of the NAIC’s financial regulatory services, including the Securi es Valua on Office and the Struc-tured Securi es Group located in New York City. He provides technical assistance to the NAIC’s members and NAIC staff engaged with federal and interna onal groups with an interest in the state-based system of fi-nancial solvency oversight, as well.

From Jan. 2008 to Jan. 2012, Mr. Johnston served as the President of the insurance regulatory consul ng firm of Rector & Associates, Inc. In that role, he provided consul ng services to insurance com-panies, insurance regulators, law firms, accoun ng firms, and oth-ers with respect to a wide variety of insurance regulatory issues. For 15 years, Mr. Johnston has held various senior posi ons with the NAIC, including Chief Financial Officer and Director of Financial Repor ng & Analysis. Jeff spent the early part of his career with the Kansas Insurance Department and Employers Reinsurance Corpora-

on, working in various financial posi ons. Mr. Johnston earned a B.S. in Finance from Emporia State University.

T NAIC S V O (C )


By David Keleher, Senior Insurance Specialist, Property and Casualty

I Perhaps no issue has been placed on more workers’ com-pensa on mee ng agendas in 2016 than the discussion of the Oklahoma Senate Bill 1062, also known as the Oklahoma Op on or “opt-out.” The bill, which was enacted in 2013 and made effec ve in 2014, allows “qualified employers” to opt-out of the tradi onal workers’ compensa on system. The Oklahoma Op on represents a significant development for the U.S. workers’ compensa on systems,1 which is com-pulsory for the majority of employers in all states except Texas. As such, state insurance regulators and state legisla-tors in other states want to know how the Oklahoma Op-

on is working for employers and employees. They are closely monitoring court challenges that have been ad-vanced by opponents of the legisla on and plain ff a or-neys represen ng injured employees that believe the law provides lesser benefits than tradi onal statutory workers’ compensa on in Oklahoma. Efforts are underway to a empt to introduce similar legisla-

on in Tennessee, South Carolina and Georgia. At the NAIC 2016 Spring Na onal Mee ng, representa ves in Tennessee and Georgia presented on their efforts to introduce the Ok-lahoma Op on in their states and the Interna onal Associa-

on of Industrial Accident Boards and Commissions (IAIABC) handed out informa on from the pending release of their study on the Oklahoma Op on. Mr. William Minick, repre-sen ng the Associa on for Responsible Alterna ves in Workers’ Compensa on (ARAWC), discussed the organiza-

on’s efforts to educate the public about the benefits of opt-out legisla on. He also explained ARAWC efforts to promote opt-out legisla on in addi onal states. Moreover, a repre-senta ve from Oklahoma discussed some of the court chal-lenges to the opt-out approach and efforts in the Oklahoma legislature to “clean-up” aspects of the 2013 Bill that had created cons tu onal challenges. W O O ? The Oklahoma Op on allows employers with sufficient size and financial resources to provide medical and indemnity benefits (wages for lost me) to workers who are injured in the course of employment. As implemented in Oklahoma, the employer would retain immunity from lawsuits by em-ployees injured in the workplace. Under the Oklahoma Op-

on, the employer would provide benefits equal to, or be er than the statutory benefits enacted in the Oklahoma legislature, including payments for medical bills, compensa-

on for lost wages and rehabilita on expenses.

Since opt-out plans are not workers’ compensa on bene-fits, they are touted as being regulated by the Employee Re rement Income Security Act (ERISA) rules, rather than by state workers’ compensa on rules. Proponents of the opt-out approach say this causes employers to communi-cate plan coverage to employees be er than workers’ com-pensa on coverage that few employees understand. Proponents of the opt-out approach point to very large re-duc ons in the cost of benefits to injured employees as a great reason to support the legisla on. However, there are several addi onal reasons large employers develop opt-out plans. These include: • Removal of claims not truly work related; • Possible reduc on in the length of temporary

disability claims; • Reduc ons in medical expenses; • Greater employer control of medical care • More mely return to work; • Prompt closure of claims; • Ability for the employer to define what injuries are

covered; • Strict claims repor ng obliga ons for employees; • Ability to direct treatment op ons; and • Employer control of an appeal process. C C There have been a number of court challenges to the Okla-homa Op on. Most of these do not stem from the size or dura on of benefits provided by opt-out plans, but from provisions contained in some opt-out plan documents that make it difficult for an injured worker to collect benefits if he/she does not comply with some very rigorous rules. One example in many opt-out plans is the “immediate report of injury provision.” If the injury is not reported within the standard me allowance, regardless of reason, the employ-er can deny the claim. Another common restric on is the length of service re-quirement. If the employee has worked for the company less than 180 days he/she cannot file a claim for a repe -

ve injury like carpal tunnel or trigger finger. Some opt-out plans contain provisions permi ng the employer to defer permanent par al disability for workers who eventually return to their jobs. These provisions have sparked a num-ber of lawsuits contending certain provisions of the opt-out are uncons tu onal.


T W ’ C O -O D C

1 Every state has its own set of laws and regula ons. Most require employers to have a workers’ compensa on policy to cover costs such as lost income or medical bills should something happen to an employee while at work.


T W ’ C O -O D C (C )

Recent Oklahoma Court Decisions • On Mar. 1, 2016, the Oklahoma Supreme Court in

Torres vs Seaboard Foods LLC. (Case No. 113649) ruled the provision in the Seaboard Foods opt-out plan, which prohibits workers’ compensa on claims from workers who have been employed less than 180 days, is uncons tu onal. In this case Torres, who sustained cumula ve injuries, had been employed only 120 days.

• On Feb. 26, 2016, the Oklahoma Workers’ Compensa-on Commission determined a provision in the opt-out

por on of the state’s 2013 workers’ compensa on stat-ute is uncons tu onal. In Vasquez v. Dillards, Dillards denied Vasquez’s shoulder and neck injury claim saying her injury was a pre-exis ng condi on and not an injury as defined by the Dillard’s opt-out plan. The Workers’ Compensa on Commission ruling suggests opt-out plans create a dual-system under which injured work-ers are not treated equally. Proponents of the opt-out approach will likely appeal this decision to the Oklaho-ma Supreme Court.

• On Apr. 12, 2016 the Oklahoma Supreme Court ruled a provision in the Oklahoma Op on, authorizing deferral of payments for permanent par al disability for work-ers who eventually return to work, is uncons tu onal. Under this provision an injured employee who returns to work receives no compensa on for the physical inju-ry sustained and no compensa on for a reduc on in future earning capacity.

T R M -R S Morantz/Kelso Study A study by Alison D. Morantz, a James and Nancy Kelso Pro-fessor of Law at Stanford Law School, analyzed the frequency and severity of claims data for 15 large mul state companies that operate homogenous facili es na onwide and com-pared outcomes in tradi onal workers’ compensa on versus opt-out workers’ compensa on for each company for the years 1998 to 2010.2 Par cipa ng employers included nine retail chains, three manufacturers and three service firms. The study found: • These companies using an opt-out approach saw a 44%

drop in costs per worker.

• More serious claims involving replacement of lost wag-es are about 33% less common among employers using opt-out plans vs. tradi onal statutory plans.

• Large nonsubscribers typically set up benefit programs to cover their injured workers.

• Benefit programs under the opt-in approach in Texas3 are not subject to exclusive remedy provisions. That allows injured workers to bring tort li ga on outside the workers’ compensa on system.

• The frequency of non-trauma c injuries declined about 47% with the opt-out plans.

In an ar cle reported by Stephanie Goldberg in Business Insurance, Ms. Morantz said “It could be that nonsubscrib-ers are be er at screening out non-trauma c claims under one of the many exclusions that private plans typically con-tain,” no ng carpal tunnel syndrome and asbestosis are rarely covered by alterna ve benefit plans. Ms. Morantz also noted, “A non-trauma c injury that is cov-ered might be denied if it’s not reported by the end of an employee's shi or within 24 hours.” However, the study showed 13 of the 15 employers have “good cause” provi-sions allowing a claims administrator to determine if there was a good reason the injury claim was made late. The research seems to indicate that not work-related inju-ries may be more quickly iden fied by the more intensive screening process used by opt-out compensa on plans. The lower frequency of non-trauma c claims could be due to either of these factors, or some combina on of both. The study did not address the welfare of workers. It’s diffi-cult to compare companies that opt out with companies that don't “because you can easily end up comparing apples to oranges,” Ms. Morantz said, adding, “If I found an oppor-tunity to carry out a really rigorous study on that ques on, I would probably go for it.” ARAWC Comments on the Morantz Study During his remarks at the NAIC 2016 Spring Na onal Mee ng, Mr. Minick (ARAWC) ci ed the cost saving for the 15 employers in the Morantz study. He men oned this study, in addi on to research conducted by the brokers repre-sen ng the 60 companies in Oklahoma who have elected opt-out plans, are experiencing tremendous cost savings.


2 “SLS Professor Alison Morantz Finds Companies Cut Costs by Op ng Out of Work-ers’ Compensa on,” March 21, 2016. h ps://law.stanford.edu/press/sls-professor-alison-morantz-finds-companies-cut-costs-by-op ng-out-of-workers-compensa on/

3 Texas is not an opt-out state. Conversely, Texas may be considered an “opt-in” state because workers’ compensa on coverage is not required of employers there. They can choose to buy insurance, or not.


Interna onal Associa on of Industrial Accident Boards and Commissions (IAIABC) Study The IAIABC is an associa on represen ng the interests of its members in the U.S. and other countries around the world. It provides informa on and so ware to members who actu-ally have administra ve responsibility for the handling of workers’ compensa on claims. The IAIABC contracted with a former Execu ve Director and insurance regulator, Mr. Greg Krohm, to conduct a study of opt-out plans in Oklaho-ma, Tennessee and South Carolina. The recently released IAIABC study4 reviewed 50 opt-out plans from a variety of Oklahoma employers and compared the benefits provided in these plans to the statutory bene-fits employees that would have been provided in tradi onal workers’ compensa on plans.

The en re study is over 60 pages and all of the findings are too voluminous to report in this ar cle. However, Figures 1 and 2 reflect a few of the key findings. The study concludes: • Opt-out plans were first developed in Texas, in a very

different regulatory environment than other workers’ compensa on systems.

• Considerable variability exists between opt-out plans that make them difficult to compare to tradi onal workers’ compensa on systems.


T W ’ C O -O D C (C )

F 1: E W ’ C ? Oklahoma “Opt-Out” Op on

Benefit Levels 70% of weekly wage up to 70% of State Average Weekly Wage (SAWW). Benefits are tax free.

Variable but most enumerate Tempo-rary Total Disability (TTD), Permanent Par al Disability (PD), and Temporary disability (TD). Some plans pay up to 100% of wage loss. Benefits are taxable.

Medical Treatment

Employer has ini al choice of provider with limited ability to change if re-quested. Fee schedule, treatment guidelines, formulary.

Variable. Dependent on individual plan and claims administrator prac c-es. Ability to direct all medical care.

Coverage Limita ons

“injuries arising out of the course of employment,” with some excep ons. Repe ve mo on excluded un l reaching 180 days of employment. (This provision recently ruled uncons -tu onal.)

Variable. Most are similar to workers’ compensa on statute. Some contain exemp ons for repe ve mo on, asbestos exposure, silica, mold, or parking lot injuries.

Disputes

Informal media on available with an ALJ. De Novo hearing before an ALJ, Appeal to Workers’ Compensa on Commission, and final Appeal to Okla-homa Supreme Court.

Variable. Internal appeal process defined by plan. Appeal to Workers’ Compensa on Commission in a trial de novo.

Regulatory Oversight

Workers’ Compensa on Commission responsible for enforcing compliance with coverage and benefit require-ments. Collect claims data. Approves self insurers. Administers dispute sys-tem. Insurance Commission approves rates and regulates workers’ compen-sa on insurers.

Insurance Commission responsible for approving plans and monitoring solvency. No further repor ng re-quirements.

Oklahoma Tradi onal Workers’ Compensa on

4 The study, tled “ IAIABC Forward: Understanding the Opt-Out Alterna ve,” is available at: h ps://www.iaiabc.org/images/iaiabc/Resources/Understanding-the-Opt-Out-Alterna ve_05-11-2016_Final.pdf.

Source: IAIABC.


T W ’ C O -O D C (C )

• The Oklahoma Op on was recently declared uncon-s tu onal by the Oklahoma Workers’ Compensa on Commission. Con nued liga on may drama cally influence this landscape.

Readers are encouraged to contact the IAIABC and re-quest a copy of the complete study when it is released. Jennifer Wolf Horejsh, Execu ve Director, IAIABC at [email protected]. C Regulators will con nue to receive inquiries from busi-ness leaders and state legislators interested in the poten-

al cost savings opt-out plans may produce. The sample size of studies conducted so far is too small to say with any certainty that opt-out plans are any more cost effec-

ve than well-run self-insurance, large deduc ble or oth-er loss sensi ve workers’ compensa on ra ng plans. The NAIC Workers’ Compensa on (C) Task Force has been charged with the task of monitoring the success of opt-out plans and will con nue to research the issue and report them at NAIC mee ngs and through the CIPR Newsle er.

A A

David Keleher is a Senior Property and Casu-alty Specialist at the NAIC. Dave has over 46 years of experience in the commercial insur-ance field and has worked as an insurance company regional manager, agency owner, insurance broker and risk management consultant. In his present posi on, Mr. Kele-her provides advice and counsel to insur-

ance regulators in all 50 states, the District of Columbia, and five U.S. territories. Mr. Keleher is a past president of the Kansas City and Los Angeles Chapters of the Chartered Property and Casualty Underwriters Society. In addi on he has a degree in risk management and has been a speaker at the Risk and Insurance Management Society annual mee ngs. He is a graduate of the University of Missouri, Kansas City with a BA degree in English/Educa on. Mr. Keleher has taught business classes at Johnson County Community College and for other organiza ons in the Midwest and Western USA.

F 2: T T D (TTD): E W ’ C ?

Length of lost me <= State Average Weekly Wage (SAWW) >= State Average Weekly Wage

(SAWW)

3 days (wai ng period) Oklahoma Op on may be superior because plans do not have wai ng

periods.

Oklahoma Op on is superior because plans do not have a wage cap, as in

workers’ compensa on.

5-21 days Ambiguous. Depends on marginal tax rate of claimant and dura on of TTD. Ambiguous. Depends on SAWW and

plan benefit caps.

>21 days (retroac ve)

Workers’ compensa on superior be-cause wai ng is eliminated. Opt-out benefits are taxed, producing lower

overall benefits.

Ambiguous for workers earning well over the SAWW.

Source: IAIABC.


By Sara Robben, Sta s cal Advisor The number of U.S. insured catastrophe losses have been below average in recent years. The severity of catastrophic events in 2011 and 2012 caused an increase in the amount of insured losses, as well as an increase in homeowners direct loss ra os, which is reflected in Figures 1 and 2. Con-versely, the years 2013, 2014, and 2015 were less than av-erage for severe catastrophic losses, which is also reflected in the figures. This ar cle features data for direct loss ra os for the homeowners’ lines of business and discusses the impact of catastrophes on those loss ra os. Insured losses resul ng from catastrophes resulted in roughly $16.1 billion dollars in 2015. (Figure 2.) Severe thunderstorms caused $9.6 billion dollars in losses, and winter storms caused $3.5 billion in losses. In comparison, catastrophes occurring in 2014 had roughly $15.5 billion dollars of insured losses, with most of the loss-related natu-ral disasters due to weather events; namely winter storms which caused an es mated $2.3 billion in insured losses, and severe thunderstorms and tornadoes losses that to-taled $12.3 billion dollars. Catastrophic losses in 2013 to-taled $13.1 billion dollars insured losses and were also largely due to severe thunderstorms and winter storms. Hurricane Irene occurred in 2011, as well as several wild-fires, tornadoes and winter storms. While there were fewer catastrophic events in 2012, Superstorm Sandy did cause damage; however, it was not as severe as the damage that occurred in 2011.

The NAIC publishes market share reports for various types of insurance. For market share informa on regarding other lines of business, the 2016 Property and Casualty Market Share will be published in mid-2016. Once published, the can be found on the Research and Actuarial Department website at www.naic.org/research_actuarial_dept.htm.

D G – U.S. I C L

F 1: H L R A A

Sara Robben is a Sta s cal Advisor at the Na onal Associa on of Insurance Commis-sioners (NAIC). She has worked in the Re-search and Actuarial department of the NAIC for the past eight years. Her current projects include staff support for the Cyber-security (EX) Task Force, the Catastrophe Response (C) Working Group, the Catastro-

phe Insurance (C) Working Group, and the Transparency and Readability of Consumer Informa on (C) Working Group. Ms. Robben worked for AIG early in her career as a Claims Adjuster and later as a Financial Analyst. She also worked on the technical side as a LAN Administrator and Technical Trainer. Ms. Robben taught technology courses for DeVry University for ten years including cours-es in computer networking, web architecture, web page design, data-base administra on, programming, and network and opera ng sys-tems security. Ms. Robben holds a bachelor’s degree in Mathema cs and Sta s cs, and a master’s degree in Project Management.

F 2: U.S. I C L ($ , 2015 )

Source: The NAIC 2015 Property and Casualty Annual Statement Data.

Source: PCS/ISO.


By Daniel R. Robles, PE, MBA, founder of The Ingenesist Pro-ject and chairman of the FinTech Task Force for the Na onal Society of Professional Engineers*

*The views expressed in this ar cle are the opinions of the author. This ar cle is not meant to represent the official posi on of the NAIC or its members.

I The purpose of this ar cle is to introduce the reader to blockchain technology and to explore how it may be used to assist with iden fying and facilita ng risk transfers in the insurance industry. Blockchain technology carries a promise of great opportunity and efficiency in business opera ons and governance. This ar cle demonstrates how blockchain technology could be integrated broadly and uniformly across society as well as the implica ons for the insurance industry. Blockchain development should not be the exclusive domain of a single sector, such as banking, nor venture-funded start-ups with ultra-high return on investment (ROI) require-ments. Likewise, purely decentralized autonomous organiza-

ons are not recommended because there is the risk of op-era ng in an extralegal sector where legal recourse may not necessarily be available when things go wrong. The primary objec ve of blockchain technology should be to reduce the cost of capital by the decentraliza on of risk. In doing so, blockchain innova on can then be applied broadly, evenly and inten onally across the economy. This makes sense because when building anything complex or important, one logical piece needs to go in front of the next logical piece regardless of its individual ROI, because the collec ve ROI is the true basis of valua on. If people tried to build an airplane in the same manner we are now trying to build decentralized economics, a few may benefit, but an air transporta on system, as a whole, would be severely constrained. This ar cle suggests the place to begin developing block-chain technology is through a consor um of insurance and professional engineering en es in the crea on of relevant infrastructure and its deriva ves upon which everyone de-pends. This includes renewable energy, clean air, water, transporta on systems, health and welfare, building sys-tems, computer networks, etc. A er all, bitcoins are not worth a whole lot when the power goes down. Infrastructure projects, and all their beneficiary deriva ves, require financial ins tu ons that can bridge the gap be-tween the incep on of a project and revenue from the pro-ject. This period of me is rife with peril because the “money and tle” precedes the delivery of the physical as-

set. The cost of capital is directly propor onal to the risk associated with project delivery. Wherever the insurance industry is capable of pooling project risks, the cost of capi-tal is greatly reduced. The insurance industry is an impera-

ve component to this objec ve. This idea represents both the challenge and the opportunity facing insurance and engineering industries related to block-chain technology. In order to arrive at these ideas, the fol-lowing paper is organized into roughly 3 parts. • Part 1 answers the ques on: What problem does block-

chain solve? This ar cle begins with a brief history of databases and draws the connec on to how society organizes itself around technology and why organiza-

onal incen ves are important to insurance.

• Part 2 suggests if each component part of the block-chain system is insurable, so too should the en re sys-tem. The insurability of the individual components of a blockchain ecosystem, revealing a somewhat mixed outcome will also be discussed.

• Part 3 iden fies how the insurance industry and the professional engineering industry together can bridge the capitaliza on gap in blockchain system insurability.

Taken together, insurance and engineering may have a pro-found opportunity to create a hybrid path crossing the digi-tal-to-physical divide for everyone to cross, including banks, venture capitalists, entrepreneurs and decentralized auton-omous organiza ons, to far greater financial benefit than a emp ng to go it alone. P O : W P D B S ? The Block Chain Protocol is a brilliant set of ideas that can-not be uninvented. Blockchain is here to stay and will likely appear in many forms and adapta ons with or without the so-called crypto-currencies that o en receive the most me-dia a en on. As we enter the next phase of blockchain development and adop on, it is now recognized that bitcoin (lowercase b), as a currency, has several flaws con-

nuing to limit its ability to completely replace money as we know it. However, Bitcoin (upper case B), as a protocol for the exchange of value, will likely remain an extremely important innova on that will con nue to be adapted in many forms. The main problem blockchain solves results from the fact computer databases simply cannot talk to each other with-out a layer of expensive fault-prone human administra on


I : T H B U C B T


I : T H B U C B T (C )

or bureaucracy. Blockchain technology is a new so ware architecture providing shared, immutable records—making processing transac ons less error-prone. This so ware ena-bles process efficiency, as well as organiza onal efficiency. Blockchains may apply everywhere people interface with a computer database. It is easy to envision the magnitude of this poten al. Before Bitcoin, if a person sent a contract over email, each party would hold an easily manipulated iden cal copy. A er Bitcoin, a person can send a contract electronically, and the receiving party would hold the only valid copy. While this may sound trivial at first, it is extraordinarily difficult for a computer to perform. But to accomplish this would, in effect, allow computers to perform some, but not all, of the administra ve func ons administrators rou nely perform today at nearly every interac on with a computer. Not unlike mechaniza on of the last century, once achieved, the so ware-administered management will be faster, more reliable and cheaper, while the marginal cost of adding addi onal capacity approaches zero. Blockchain may scale up to handle large and complex transac ons or scale down to accommodate billions of micro-transac ons. Also like mechaniza on, society will certainly reorganize around these new forms of value crea on and exchange. This is already evident with the extraordinary amount of venture and investment capital and crea ve new decentral-ized autonomous organiza ons pouring into blockchain space. The technology required to accomplish this is complex, and un l Bitcoin, it was not obvious how such a thing could be accomplished at scale. Bitcoin taught the world a great deal with its success and its limita ons. However, the fact re-mains it is extremely difficult for a computer to make hu-man-like decisions, and the full breadth of human capability is s ll well outside the reach of blockchains. Meanwhile, the poten al risk-sharing partners in this technology have not yet determined the full implica ons, nor even created the effec ve vocabulary to describe its switches, knobs and pedals—let alone the blocks and chains. Technology Reorganizes Society The importance of blockchains includes their impact on how people may reorganize in a community. For insurance com-panies, the organiza on (mutualiza on) of likely perils is a core aspect to correct pricing of insurance products. To un-derstand these implica ons requires a brief history lesson on database architecture.

In the early days of computer networks, machines that per-formed computa ons were connected with wires to other machines that stored data on some physical medium such as magne c tape. Humans interacted with these machines by using finger symbols (keyboard) and changing reels of magne c tape. These ac vi es had very li le to do with the computa on actually being performed. While we may not realize it, those same func ons are s ll o en performed today in one form or another every me we interact with a computer database. Over me, databases became so incredibly useful that com-panies and ins tu ons stored all of their data in proprietary silos, where they could control access to financial records, product specifica ons, trade secrets, personnel files, cus-tomer data and sales projec ons, etc. The database for an aircra manufacturer was structured en rely differently from a database for a coffee shop chain or an insurance company. The specialized linkages formed between the data and the opera ons became unique to the organiza on and, in many cases, proprietary. The purpose of management was to let nothing in or out of the database without permis-sion. It has been widely wri en how ins tu ons have be-come defined by their data structures. The problems with legacy databases became apparent when the need arose for one database to communicate directly with another database. But this was impossible without hu-man administra on. With the advent of the Internet and social media, widespread networking capability between computers (nodes) became exponen ally more valuable, while the ability for computers to communicate with each other remained linear. While electrons moved at the speed of light, many systems remained limited to the speed of bureaucracy. In the 1990s, organiza ons introduced legions of adminis-trators, intermediaries and brokers to help databases com-municate with each other. More recently, database engi-neers invented special interfaces (APIs) that allow, say, Am-azon.com to provide access to parts of their database to wholesalers or partnered retailers. APIs allowed for a wave of innova on associated with the ecommerce movement and much more. However, even APIs had significant short-comings with more formal transac ons. For example, with all the APIs in the world, a real estate broker in 2016 must s ll wrestle with several databases in order to complete a transac on. They must lead the buyer and seller around the mul ple lis ng service database



(MLS), as well as coordinate a lender, property inspectors, property insurance, escrow service and tle insurance—all under strict government regula on and management over-sight. The agents must deliver all of these databases in rela-

ve unison to a single point in me to receive signatures, a “ me stamp,” and become registered in public archives. And, the deal can s ll be reversed by a legal challenge. The process can take weeks or months, with unnerving cost fric ons and price vola lity. “This is all very weird, only we’ve become accustomed to it.” — Vinay Gupta Unfortunately, as the value of data increases, so too are the incen ves, probability and the consequences of chea ng, especially where the ability to cheat has been equally en-hanced by the same imperfect technologies. Addi onal laws and regula ons are some mes applied, which may thwart innova on to a greater degree than the protec on those laws may provide. Today, asymmetric informa on, blanket legisla on and selec ve enforcement are considered among the scourges of modern day commerce. Keep in mind much of this has very li le to do with the actual thing that is try-ing to be accomplished. What if we can get rid of all of those hindrances? What if we can eliminate the brokers and intermediaries and the bureaucracy and the administra on and the noise and the fric on, etc? Actually, this is a popular idea a empted throughout histo-ry in various forms of governance and marked by the will-ingness and ability to control informa on. Obviously, there are many methods for applying control (or not applying control); most lay on a spectrum between a fully centralized organiza on and a fully decentralized organiza on. The benefits and drawback of each are well understood from historic references—that is, un l blockchain technology arrived. Centraliza on The first way to enable databases to communicate with each other is to consolidate and combine them into a single database hoping enough commonality would exist to patch them both together. These are aptly called “acquisi ons and mergers” where two somewhat similar en es com-bine their data under a central authority. Efficiencies are gained in scale and elimina on of redundancy. Unfortunate-ly, centraliza on can also lead to inefficiencies such as top-heavy hierarchy, monopoly, obfusca on, stagna on and vulnerability to groupthink or external shocks. Failures

would o en trigger blanket legisla on. Meanwhile, the orig-inal problem remains: How do these new mega databases communicate with other mega databases? Decentraliza on The other way to eliminate intermediaries is for everyone to share the same database between organiza ons. Mul -ple writers can retrieve and populate data simultaneously with no controls, consensus or centralized authority. Natu-ral organic linkages would form and opera ons would be-come faster, cheaper and easier to perform and maintain. The network effect can take hold where the value of the network would grow exponen ally. Unfortunately, there would be no way to stop a person from chea ng another person, or going back to change the condi ons of a con-tract, or giving himself a raise or double spending a unit of account, etc. For decentralized databases, these are pre-cisely the problems blockchain solves. What Does This Mean for Insurers? People and organiza ons will reorganize around this new type of data and value exchange system much like they ear-lier reorganized around prior technologies, such as typing pools. This represents a new set of business perils that do not necessarily pool well with the old set of business perils upon which current insurance products are based. In es-sence, the insurer is faced with four primary concerns. • How different would it be to insure a decentralized

business or business processes than a centralized set? What historic data are s ll valid? What data needs to be collected anew? How much can the insurer rely up-on a management system comprised of nothing but so ware? How does an insurer assert dominion over economic value denominated in cryptographic tokens that are neither money nor property according to the law? Who do you call when things go boom?

• The insurance industry itself is an administra on-laden database. Could it operate on a blockchain? What are the opportuni es and implica ons of culling their own legions of brokers and staff? Would an insurer be will-ing to insure a company that had just culled their own brokers and staff? If they do not do it and a compe tor does, what perils are then imposed on the firm? How does the insurer preserve ins tu onal knowledge in the wake of replacing brokers with so ware?

• The purpose of regula on of any kind should be to en-courage or discourage specific types of human behav-iors. If the human is taken out of the equa on, what regula ons are s ll needed? Are there any regula ons





standing in the way? Are new regula ons required? Can regula ons be bypassed or shi ed to another seg-ment of a process? How fast can regulators respond to an unan cipated condi on?

• Finally, everything about database management has very li le to do with the thing actually being computed. Blockchain and crypto-currencies exist in a digital realm. Meanwhile, real people are doing real things in real life where real things behave according to physical laws. How exactly will blockchain so ware reconcile or interact with the real world?

These are extremely important ques ons yet to be re-solved. It is worth the me and effort to learn and under-stand the implica ons of blockchain technology because the opportuni es for adop on by the insurance industry are quite literally exponen al: • Insurers may achieve efficiency with internal processes.

• Insurers may achieve efficiency insuring blockchain clients.

• Insurers may discover new markets previously unviable.

• Insurers may reduce the granularity of insurance prod-uct to ghter pools.

• Insurers may scale up or scale down (micro-insurance) at near zero marginal cost.

The insurer needs to know exactly what is being insured, the numerical probability the peril will or will not manifest, and the consequences of a failure or breakdown in the pro-cess. Problems may arise where an organiza on loses im-portant ins tu onal knowledge, adaptability and innova-

on due to the wholesale elimina on of important adminis-tra ve personnel. The second part of this ar cle will dissect a blockchain pro-cess into five cons tuent parts and analyze the insurability of each subsec on. If all segments of a business process are insurable, then the en re process ought to be insurable. This second part will demonstrate how exis ng ins tu ons may help bridge insurability gaps in blockchain implementa ons. P T : T M B Our theory is if each component part is insurable then the en re ecosystem should be insurable. Using a simple insura-bility test, we can iden fy shortcomings of a business plan needing to be shored up with non-blockchain ins tu ons, or we know the plan is unviable. Further, blockchain applica-

ons that are the most durable from an insurability stand-

point, may also signal the best returns on blockchain invest-ment and enjoy lower cost of capital for funding innova on. The Insurability of Blockchains Investment in any innova on or asset requires ins tu ons willing to carry the cost and risk of design, development and construc on of a project before—some mes years before—the asset produces revenue sufficient to return the investment capital. The cost of capital is o en the primary driver determining what can and what cannot be built. Where an investment can be insured, the cost of capital drops precipitously. Blockchain technology is like a three-trick pony. It essen ally combines three slightly clumsy computer tricks in order to mimic decisions a human administrator rou nely makes with apparent ease. The difference is, if done correctly, the computer can perform some of these decisions with great speed, accuracy and scalability. If done incorrectly, the com-puter can also propagate an incorrect outcome with stun-ning efficiency. The technique we will use to analyze insurability harks back to any “Insurance 101” textbook with the three condi ons of insurability expressed as follows: • Can we iden fy the risk exposure?

• What is the (mathema cal) probability such risk expo-sure will manifest?

• If so, what are the consequences (cost) of failure?

The rules of our test are simple: All three condi ons must be known in order to create an insurance product. The inability to answer any one of these ques ons results in a non-insurable condi on. Non-insurable business methods using blockchain technology must then be augmented or rejected. #1: The Byzan ne General’s Dilemma This scenario was first described in 1982 at SRI Interna onal. This problem simula on refers to a hypothe cal group of military generals, each commanding a por on of the Byzan-

ne army, who has encircled a city they intend to conquer. In formula ng their plan, it is determined there are only two ways to win the war: 1) they all must a ack together or 2) they all must retreat together. Any other combina on would result in complete annihila on. Obviously each general has a vested stake in the outcome of the group’s consensus. The problem is complicated by two condi ons: 1) there may be one or more traitors among the leadership working for



the other side; and 2) the messengers carrying the votes are subject to being intercepted. For instance, if a traitorous general could send a e-breaking vote in favor of a ack to those who support the a ack and a different vote to those who support a retreat, a rout could be inten onally and easily created. A Byzan ne fault-tolerant system may be achieved with a simple test for unanimity. A er the vote is called, each gen-eral then “votes on the vote,” verifying that his own vote was registered correctly. The second vote must be 100% unanimous. Any other outcome would trigger a default or-der to retreat. Metcalfe's law states that the value of a telecommunica-

ons network is propor onal to the square of the number of connected users of the system (n2) — Wikipedia. Metcalfe’s law provides a means of predic ng the security of such a network would also be propor onal to the square of the number of members in the network. Insurability Test #1 Next, we apply the three condi ons of insurability to the above scenario: • Yes, we can iden fy the risk exposure to the generals

and their armies.

• Yes, ci ng Metcalfe’s law for networks, the probability of corrup ng the network may be readily calculated using real numbers.

• Indeed, the consequences of failure would be tragic, but at least an insurance product could be offered to the families.

Therefore, #1 is insurable. Modern Examples of Byzan ne Fault Tolerant Systems The analogy for networks is that computers are the gener-als, and the instruc on “packet” is the messenger. To se-cure the general is to secure the system. Similar strategies are commonplace in engineering applica ons, from aircra to robo cs or any autonomous vehicle where environmen-tal inputs are converted to movements of, say, a flight con-trol surface. The Boeing 777 and Boeing 787 use Byzan ne proof algorithms, and each are clearly insurable mecha-nisms in a highly regulated industry of commercial avia on. #2: Mul key Cryptography The Byzan ne fault-tolerant strategy is useful for securing the nodes in a network (the generals). Mul key cryptog-raphy is for securing the packets of informa on they ex-

change between them. On a decentralized ledger, it is im-portant that the people who are authorized to access infor-ma on and the people who are authorized to send the in-forma on are secured. It is also important the informa on cannot be tampered with in transit. Society now expends a great deal of energy in bureaucra c systems performing these essen al func ons to prevent the , fraud, spoofing and malicious a acks. Trick #2 allows this to be done with so ware. Assume for a moment that a cryptographic key is like any typical key for opening locks. The computer can fabricate sets of keys that recognize each other. Each party to the transac on has a public key and a private key. The public key may be widely distributed because it is indiscernible by anyone without the related private key. Suppose Alice has a secret to share with Bob. She can put the secret in a li le digital vault and seal it using her private key + Bob’s public key. She then sends the package to Bob over email. Bob can open the packet with his private key + Alice’s public key. This assures the sender and receiver are both authorized and the package is secured during transit. Insurability Test #2 Applying the three condi ons of insurability : • Yes, we can see the risk exposure to an unsecured mes-

sage.

• Yes, we can calculate the probability of failure by exam-ining the strength of the encryp on.

• Indeed, the consequences of failure could be es mated because the contract would likely represent value.

Therefore, #2 is also insurable. Things are looking up. #3: The Time Keeper Einstein once said that the only reason for me is so every-thing does not happen at once. There are several ways to establish order in a set of data. The first is for everyone to synchronize their clocks rela ve to a small borough of Lon-don and embed each and every package with dates of crea-

on, access, revision and date of exchange, etc. Then we must try to manage these individual posi ons, revisions and copies moving through digital space and me. The other way to accomplish this is to create this moving background (like they do in the old TV cartoons) and indelibly a ach the contracts as the background passes by. In order to corrupt one package, you would need to hijack the whole





train. The theory is that it would be prohibi vely expensive, far in excess of the value of the single packet, to do so. Computer so ware of the blockchain performs the follow-ing rou ne in order to accomplish the effec ve equivalent process: Consider for a moment a long line of bank vaults. Inside each vault is the key or combina on to the vault im-mediately to the right. There are only two rules: 1) each key can only be used once; and 2) no two vaults can be open at the same me. Ac ng this out physically, it is a bit of a chore, but security is assured, and there is no way to go backwards to corrupt the earlier frames. The only ques on now is: Who is going to perform this chore for the benefit of everyone else and why? Finally, Here is Why the Coin is Valuable There are several ways to push this train along. Bitcoin uses something called a proof-of-work algorithm. Instead of hid-ing the combina ons inside each vault, a bunch of comput-ers in a worldwide network all compete to guess the combi-na on to the lock by solving a puzzle that is difficult to crack but easy to verify. It is like guessing the combina on to a high school locker. It is hard to do, but once accomplished, everyone can easily see the open locker; that is sufficient proof the work has been done. Whoever solves the puzzle is awarded electronic tokens called bitcoin (with a lower case b). This is sort of like those li le blue ckets that kids get at the arcade and can be ex-changed for fun prizes on the way out. These bitcoins simp-ly act as an incen ve for people to run computers solving puzzles to keep the train rolling. Bitcoin (All Crypto Currencies) Must Have Value Because If They Did Not, Their Respec ve Blockchain Would Stop Cold A stalled blockchain would be the crypto-currency equiva-lent of bankruptcy. This may account for a fair amount of hype around the value of bitcoins. Not surprisingly, as the price increases, the be er the blockchain operates. Insurability #3 While all of this seems a bit confusing, keep in mind that we are describing the thought pa erns of a computer, not nec-essarily a human. The important thing is that we can ana-lyze the mathema cs: • Yes, we can see the risk exposures associated with

vaults, trains and puzzles.

• Yes, we can calculate the probability that the system can be corrupted by the rela ve value of the coins.

• Indeed, the consequences of failure could be dire, but the hazards are foreseeable.

The Blockchain Ecosystem So there we have it. All three are insurable and, therefore, we can say with ra onal confidence blockchains are insura-ble for their intended outcome. The problem is blockchains cannot exist in digital isola on; their value must be derived from the value of something else—something real. Are Cryptocurrencies Actually Money? There are many prominent ar cles by many smart people discussing this topic. However, at the me of this wri ng, according to Uniform Commercial Code, ar cle 9, a very explicit defini on for money is provided as follows: "Money" means a medium of exchange currently author-ized or adopted by a domes c or foreign government. In terms of our insurability test, the answer is simple: No, digital tokens are not money. While their destruc on may represent an economic loss, the loss would need to be de-nominated in dollars. The courts and enforcement cannot be invoked to protect your bitcoin. While we may be able to iden fy the peril and even calculate the probability of loss, we cannot predetermine the consequence of the loss and, therefore, cannot price the risk correctly. Are Cryptocurrencies Considered Property? There is some ambiguity here as well. When we think of prop-erty, we think of discreet units that are largely inseparable. The tle to the asset travels with the whole asset as it chang-es hands. A lien on the property would be needed in order to assert dominion on the asset. But bitcoins are quite easily divisible, almost fluid, lubrica ng a blockchain. If I loaned you a car but kept the wheels as collateral, the u lity of the car would be encumbered. Or it would be like holding a lien against the money to purchase the car—and not the car. The answer for all prac cal purposes is that cryptocurren-cies cannot really be treated as property, at least within the boundaries of law. Therefore, they are uninsurable. Let Us Take a Look at Where We Are


Insurable #1 Fault Tolerant

#2 Mul key Cryptography

#3 Decentralized Ledger

Represents Money X Represents Property X


So, if bitcoins are not money and bitcoins are not property, what are they? How does one prove ownership? How does the owner assert dominion? How would liability be assigned for economic losses of another person in a transac on where all agreements are in the form of nonrevocable con-tracts executed by so ware? Where do rights and responsi-bili es a ach? This is a deeply troublesome discussion if you are in the business of assuring or insuring blockchain-based enterprises. More troubling is that these precise characteris cs are what make cryptocurrencies a rac ve for illegal ac vity, thereby increasing variance of expecta ons rather than reducing variance—the exact counter-effect of insurance. If assets can be converted to cryptocurrency, they become difficult to seize or repossess. The extra-legal sector is categorically uninsurable by mainstream carriers. The insurance industry is faced with both a dilemma and an opportunity to build specialized insurance for blockchains, or bridge the insurability gap with mainstream markets, or both. Clever legal scholars have suggested perhaps ownership may be established with a claim against the cryptographic keys that open and close the packets. This is a very inter-es ng idea. We have already established that these nodes and these keys are insurable. Logic may be built into key distribu on to assign liability or limit liability and, thus, price risk correctly. P T : B C G In Part 1 we iden fied the problems blockchain solves. In Part 2, we iden fied the problems blockchains cannot solve. In this part, we will try to specify a bridge one might build across the chasm upon which everyone from banks, entre-preneurs and modern decentralized organiza ons may cross. One alternate approach rarely considered is a hybrid mod-el of physical proofs interchangeable with the digital proofs in a blockchain, as needed or where appropriate. For example: • Instead of computer modeling a fake network of Byzan-

ne generals, a network of real generals can be set up to model a computer network.

• Instead of a solu on to a trivial puzzle as a means of genera ng a digital token, the solu on to a real life puzzle can also be used to generate a digital token.

• Instead of a hashing program genera ng a crypto-graphic key, a person’s resume could be used as the algorithm to hash cryptographic keys authorized to open and close packets on the blockchain.

• Etc.

As long as each component of the blockchain ecosystem is insurable, the en re system would remain insurable. There would otherwise be no limit to the number of blockchains that can exist nor the number or combina on of analog and digital components that can be mixed as long as the tokens, in the end, can clear accounts. The Ins tu on of Professional Engineering For 80 years, the professional engineers (PEs) have been trusted third-party adjudicators for banks and insurance companies. The prevailing purpose of PEs is to safeguard the health and welfare of people and property—and by extension, the insurers and banks and assure them. Profes-sional engineering allows people and projects to span the capitaliza on gap—that is, the me gap between the ini a-

on of investment and the delivery of revenue from the investment. The professional engineering process, in fact, provides many of the same security func ons as the three tricks of Blockchain technology. • PEs endure a peer review process in obtaining and

maintaining their license. Examina ons qualify the en-gineers and a revocable license established an incen-

ve to high integrity. This bears similarity to the Byzan-ne general’s network.

• PEs use a common science and language of mathe-ma cs as the public key and their respec ve problem solu on as the private key effec vely encoding their judgments. An engineer recognizes the informa on of another engineer and can validate the integrity of a packet of informa on. This simulates mul key cryptog-raphy.

• The PE’s stamp acts to finalize a transac on to an indel-ible legal ledger memorializing monetary value and tle to property.

The con nued similari es between the goals of blockchain protocol and the professional engineering protocol are re-markable—thus, demonstra ng individually, blockchain ideas are not new and may in fact be more compa ble to exis ng ins tu ons than previously considered. Perhaps an effec ve blockchain can be constructed combining compo-nents of each realm, real and virtual, to achieve high ambi-





guity (human capability) while also providing speed, accura-cy and scale (computer capability). The Na onal Society of Professional Engineers (NSPE) and the NAIC are at an im-portant point in history. They can either wait to see if this new technology will render important ins tu ons irrele-vant, or they can use this technology to amplify the current roles as a “financial ins tu on” in their own right. The Insurability of Engineering During the design and construc on phase of a project, the asset lives on a balance sheet only as an engineering design. The engineering stamp serves as proxy for the future tle and revenue of the project. Insurability is con ngent on engineering sign-off. If a project were not insurable, then a bank would not lend money to it. This scenario is rou ne and commonplace in modern fi-nance given the insurability of professional engineering and its ability to bridge the capitaliza on gap. Let us now take a look at where we stand on blockchain system insurability:

Oracle Contracts A “smart contract” is a decision executed by a computer algorithm on a blockchain. For example, if condi on A and condi on B are triggered, then payment C is executed. An


Insurability Matrix Blockchain Ecosystem

With Professional Engineering

#1 Fault Tolerant

#2 Mul key Cryptography

#3 Decentralized Ledger

Represent Money X

Represent Property X

C S B P PE P A ribute

Blockchain Protocol PE Protocol

Fault Tolerant

Yes Yes

Objec vity Programmed rules and computer algorithm Engineering laws and principles.

Governance Trusted third party to administer databases. Trusted third-party ins tu on to the public, banking and financial ins tu ons for 80 years

Permanence Transac ons are executed by programmed set of rules that are indelible.

Works of engineering, by nature, are irreversi-ble and indelible by observa on.

Consensus Computers that vote on the vote reach con-sensus. Mining puzzle is difficult to solve but easy to prove.

The professional engineer stamp secures the nodes by peer review. Engineering puzzles difficult to solve but easy to prove.

Chronology A string of indelible blocks establish chrono-logical order of contracts in me.

Professional engineering stamp and permi ng establish chronological order of physical state

Security Security is provided with cryptography that is very difficult to guess but very easy to prove.

Security is provided by licensure, which is very difficult to obtain/fake but very easy to prove.

Transparency

A blockchain can be audited to track cheaters or validate transac ons.

Engineering is naturally auditable. Processes track risk exposures.


adjudicated smart contract is a smart contract whose exe-cu on is con ngent on a physical observa on or judgment by a reliable witness. An Oracle contract is an adjudicated contract with the add-ed requirement the adjudicator is deemed the most appro-priate person to be performing the adjudica on. These ad-di onal requirements mean a method is required to estab-lish the most appropriate adjudicator, and the method must likewise be insurable. The Oracle must make decisions in physical space—not simply assess digital data. The Oracle must be able to be present in me and space, determine causa on of an event and deal with significant ambiguity in rela on to the facts being observed. The validity of the Ora-cle is what establishes tangibility and invokes law—therefore, money and property. Securing the pool of decen-tralized Oracles would be essen al to insurability of such contracts on a blockchain. Banks and insurance companies depend on engineers to verify the design, materials, processes, components and performance of all subjects they finance. In general, the construc on process breaks down into a long and compli-cated series of events that all must be contracted, nego at-ed, ordered in me and verified in a secure manner while also triggering payments to stakeholders. These events are

ed together by cri cal path methodology. All actuarial data used to insure any number of insurable condi ons at some point touches the professional engineering stamp. A structure cannot be occupied without the PE stamp; a car cannot be insured without safe roads and bridges; a munici-pal project cannot be capitalized without licensed engineers assessment, etc. Likewise, subsequent innova on such as autonomous mo-tor vehicles; Internet start-ups and blockchain enterprises are collec vely con ngent on safe roads, reliable computer networks and renewable energy, and are thus con ngent on insurable Oracles. Professional engineering is the best model to start with because it is already codified and insur-able. It will be essen al to broaden the breadth and depth of the Oracle pool as blockchain implementa on advances. However, the insurability requirements must remain in or-der for the global blockchain experiment to be ul mately successful. The promise of Oracle contracts is that the Oracle pool may itself be decentralized and distributed broadly across socie-ty to include a strong diversity of domain experts. As long as this process is insurable, there are no limits to who and how many Oracles may act in a network.

The Oracle would essen ally flip the switch allowing the computer to follow a path of logic to, say, approve the next step in a sequence of events; assign, limit and transfer liabil-ity; shi insurance coverage; establish responsible charge; or ini ate a payment from a bank, bond, insurance claim or con ngency fund. If there is a problem or suspected corrup-

on, the en re trail can be audited and unwound to forensic standards. C We may have been here before. Many of the issues brought up in this ar cle were also present during the me of this author’s par cipa on in the North American Free Trade Agreement (NAFTA) nego a ons. Anyone who was around in the early 1990s may remember the mantra of modern globaliza on was that decentralized markets were good and centralized markets were bad. The mathema cs suppor ng the efficiencies of the compara ve advantage economic model was, and s ll is, indisputable. Unfortunately, decen-tralized markets were administered unevenly, dispropor-

onately and were only par ally insurable, at best. The act of trying to control a decentralized market eliminated many of the benefits of having one. Today, we may face a similar peril, except with a far more powerful technology promising exponen al efficiencies or exponen al deficiencies. The difference is that we also have the knowledge, foresight and the profound responsibility to get it right this me. The consor a between engineering and insurance already exist, and their impact on the cost of capital is abundantly clear. To formalize this on a blockchain ini a ve is not a radical posi on by any means. What is unique about this proposal is that insurance and engineering should be at the forefront of blockchain development, building the bridge spaning the capitaliza on gap upon which everyone else can travel. The current path of blockchain deployment dominated by banks, venture capitalists and decentralized autonomous organiza ons is not sufficient in delivering the highest and best use for this important technology within the exis ng framework. The market incen ve and corresponding regula-tory overreach in a emp ng to control blockchains will only have the effect of re-centralizing databases rather than de-centralizing databases. The further detriment of regulatory arbitrage may serve only to increase vola lity and not de-crease it. The superior method for so-called “controlling” blockchain technology would be through hybrid applica on of digital




and physical proofing mechanisms that are individually in-surable so that their infinite combina ons would s ll result in easily insurable enterprise. Reinsurance could then be an umbrella. Unique combina ons of such components as-signed by entrepreneurs would yield the new business methods of the future at a very low cost of capital. The Oracle pools may be decentralized through algorithms conver ng resumes to cryptography in a manner that se-cures asset nodes ( tles). Real world problems can be used as proof-of-work for the puzzles that move blockchains and create their associated currency. Cryptocurrencies would no longer be just digital tokens best suited for specula on. Rather, they could represent real human produc vity achieving generalized reciprocity in real money exchanges. In the manner specified herein, blockchain technology can meet its highest poten al in delivering improved financial methods to an increasingly crowded planet.


A A Daniel R. Robles, PE, MBA is the founder of The Ingenesist Project (TIP), whose objec ve is to research, develop and publish applica ons of blockchain technology related to the technical and financial services industries. Mr. Robles currently serves as the chairman of the FinTech Task Force for the Na onal Society of Profes-sional Engineers (NSPE), as well as a research

fellow at the Interna onal P2P Founda on related to blockchain implementa ons. Mr. Robles is known worldwide as blogger for www.Ingenesist.com, www.Insurancethoughtleadership.com, www.Coengineers.com and several others. He holds professional engineering (PE) licenses in Washington and California, as well as a master’s degree in interna-

onal business from Sea le University.


© Copyright 2016 Na onal Associa on of Insurance Commissioners, all rights reserved. The Na onal Associa on of Insurance Commissioners (NAIC) is the U.S. standard-se ng and regulatory support organiza on created and gov-erned by the chief insurance regulators from the 50 states, the District of Columbia and five U.S. territories. Through the NAIC, state insurance regulators establish standards and best prac ces, conduct peer review, and coordinate their regulatory oversight. NAIC staff supports these efforts and represents the collec ve views of state regulators domes cally and interna onally. NAIC members, together with the central re-sources of the NAIC, form the na onal system of state-based insurance regula on in the U.S. For more informa on, visit www.naic.org. The views expressed in this publica on do not necessarily represent the views of NAIC, its officers or members. All informa on contained in this document is obtained from sources believed by the NAIC to be accurate and reliable. Because of the possibility of human or mechanical error as well as other factors, however, such informa on is provided “as is” without warranty of any kind. NO WARRANTY IS MADE, EXPRESS OR IM-PLIED, AS TO THE ACCURACY, TIMELINESS, COMPLETENESS, MERCHANTABILITY OR FITNESS FOR ANY PARTICULAR PURPOSE OF ANY OPINION OR INFORMATION GIVEN OR MADE IN THIS PUBLICATION. This publica on is provided solely to subscribers and then solely in connec on with and in furtherance of the regulatory purposes and objec ves of the NAIC and state insurance regula on. Data or informa on discussed or shown may be confiden al and or proprietary. Further distribu on of this publica on by the recipient to anyone is strictly prohibited. Anyone desiring to become a subscriber should contact the Center for Insur-ance Policy and Research Department directly.

NAIC Central Office Center for Insurance Policy and Research 1100 Walnut Street, Suite 1500 Kansas City, MO 64106-2197 Phone: 816-842-3600 Fax: 816-783-8175

http://www.naic.org http://cipr.naic.org To subscribe to the CIPR mailing list, please email [email protected] or [email protected]

Documents

CIPR Newsletter June 2016 › cipr_newsletter_archive › vol19.pdfData at a Glance: U.S. Insured Catastrophe Losses 16 The severity of catastrophic events in 2011 and 2012 caused