Mandatory Purchase: Bigger Than The Banks

Mandatory Purchase: Bigger Than The Banks

Written By:

Joe Rossi, ANFI, CFM
Flood Specialist, Rogers and Gray Insurance
Chair and Executive Director, Massachusetts Coastal Coalition

Tim Carty
Owner, Murphy Carty Insurance
Director, Massachusetts Coastal Coalition


Congress enacted the National Flood Insurance Program (NFIP) in 1968, intending to provide flood coverage because private insurance no longer offered it.  By 1973, only 95,000 flood insurance policies were sold nationwide. To increase the program’s acceptance, Congress enacted the Flood Disaster Protection Act (FDPA) in 1974, establishing the “Mandatory Purchase” requirement. It obliged Federally Regulated lending institutions to issue loans in high-risk flood zones only when flood insurance is simultaneously written.  The law mandated that if flood insurance is not purchased, the lender must force place coverage on the borrower.  This protected the loan from default from flooding and provided high-risk flood coverage.


It seems that every flood event brings the NFIP, and its critics, to the forefront of media attention.  With each event, such as Hurricane Harvey in 2017, come headlines telling a similar story: mandatory purchase requirement has failed.  However, we fail to separate those required to have flood insurance from those who choose not to have coverage.  Nearly 85% of Hurricane Harvey’s victims did not have flood insurance. The primary reason was because flood insurance was not required.  Mandatory purchase versus the voluntary purchase of flood insurance needs to be better understood to protect against future floods.  There are large gaps in data that cloud the uninsured issue and fail to get to the root of the problem.

The Government Accountability Office (GAO) completed the first “study” on mandatory purchase in 1990. The study, however, was inconclusive.  Following this effort, no large-scale study was attempted until after Hurricane Katrina in 2005.

Katrina was the single worst flooding event in the nation’s history. Following Katrina, the NFIP plunged into debt and concern was raised again about is effectiveness and regarding mandatory purchase.  A study done by the RAND Corporation in 2006 suggested that lender compliance was in the range of 75%-80%, but again, this report was determined inconclusive.

In 2012, an NFIP reform effort known as The Biggert-Waters Flood Insurance Reform Act was signed into law. It offered program reform, but following public outcry, it was amended by the Homeowner Flood Insurance Affordability Act of 2014. Reforms focused on flood coverage reflecting actual risk and moderating excessive premium increases.

Following the hurricanes of 2017, mandatory purchase versus voluntary purchase became a single issue. However, the two issues are separate issues.  Mandatory purchase does need better validation, but those uninsured for Hurricane Harvey, for example, will not face required coverage nor elect to have flood insurance in the future.  Homeowners in low risk flood zones tend to see weather events as anomalies. Realize that 73% of declared disasters from 2008-2017 involved flooding and only 12% of American households have flood insurance according to the Insurance Information Institute.


To fully understand the root of the uninsured issue, we must understand why people will not purchase voluntarily. This is a critical dialogue to get right.  Confusing the enforcement of mandatory purchase against the portfolio of all in-force flood policies leads to confusion on how to best protect against future floods.

Most articles criticizing mandatory purchase point to a 2014 FEMA study called “Policy Penetration and Mandatory Purchase Report.”  This report concluded that only 47% of all the structures required to have flood insurance under mandatory purchase were in compliance.

The previous head of the NFIP, Roy Wright, testified to the Financial Services Committee that the 2014 FEMA study was “not data we keep up to date, we don’t have a tracking element for it, ” pointing to inaccurate data driving flood reform. After reviewing the 2014 study, FEMA merged the discussion of mandatory purchase and total policies in force into a single category.  A critical footnote on the first page of the 2014 study summarizes that FEMA merely estimated number of policies in force, rather than determining the number of polices actually required under law.

Mr. Wright also stated in his testimony to Congress that 39% of real estate transactions in 2016 were cash-based. Cash transactions do not require mandatory purchase; flood insurance is voluntary.  FEMA’s 2014 study excluded cash transactions.  However, policy makers used the findings to make policy decisions to reform the NFIP.  In FEMA’s 2018 affordability study, they also recognized that, “In the Special Flood Hazard Area, more non-policy holders own their homes outright than have mortgages.”

Not everyone in a high-risk flood zone is required to have flood insurance under mandatory purchase. Private mortgages are regulated differently where flood insurance is not mandated. Private mortgages have more than doubled in popularity from 23.4% of all mortgages in 2008 to 48.3% in 2016.  Lender Placed insurance is typically private, accounting for approximately 10% of all mandatory purchases.

Private commercial flood must also be considered: 85% of commercial flood is insured by the private market, with no consideration in prior studies.

In determining mandatory purchase enforcement, scenarios exist where the NFIP will not provide coverage. These properties add to the uncertainty of a conclusive mandatory purchase study. A narrow number of properties exist where a building is either 100% over water, they are excluded from flood insurance due to land use regulations or are located in a non-participating NFIP community (where flood insurance is unavailable.)

With multiple studies returning inconclusive, where does the data reside?  We can start with lending organizations and their examiners.  Federally regulated lenders are examined by the Office of the Currency Comptroller (OCC), Federal Deposit Insurance Corporation (FDIC), Federal Reserve Bank (FRB) and the Consumer Financial Protection Bureau (CFBP.)


Lenders can be fined up to $2,000 per violation per loan, with an unlimited annual fine potential.  The Flood Disaster Protection Act contains up to 54 fine categories for mandatory purchase violations per loan, translating to a potential of $100,000+ in fines per loan. Annual  examinations for mandatory purchase compliance are completed with little guidance  on mandatory purchase procedures.

With a possibility of over $100,000 in violations per loan, and mandatory purchase being poorly enforced, one would expect to see egregious fine levels.  Such is not the case: in 2017, $2.8 million fines were assessed, translating to only fifty cents per flood policy if spread over 5 Million NFIP policies in force.   One major Federally Regulated lender we interviewed has 5,422,681 residential loans with 197,372 in zones that require flood insurance.  They are four years without violation and are audited annually.


The public and Congress seem to point fingers following large events regarding the number of uninsured losses.  Our point is to show that a lack of data and the lack of gathering correct data distorts the flood program’s reality.  Lenders do their job as prescribed by law.  Our focus should be on changing social norms to expand the voluntary component of flood insurance, in addition to expanding mandatory purchase coverage.

Congress proposes increasing lender fines to $5,000 per violation.  This proposal is based on bad data and will not improve the program. We need to show all households the value of flood insurance.  We need a new world of flood insurance with two types of programs: Mandatory purchase, and “Voluntary” Purchase where reasonable premiums and a sensible product will make homeowners want to purchase through NFIP, private markets, or other means. Mandatory purchase is not the solution, but is a part of a bigger solution.

*Note: Uncited statistics and numbers come from the data and information we gathered in our interviews