Gayland Kitch doesn’t feel a bit sheepish about not having a storm cellar, even though he is the director of emergency management in Moore, Okla., which faced one of the most violent tornadoes on record, with wind speeds greater than 300 mph, in May 1999.

It isn’t that Kitch is resisting the $3,000 or so it would take to build. It’s that during tornado weather, he’s not home. He’s at the office, which has its own shelter. His wife is there, too, volunteering. When their kids lived at home, they came, as well.

Kitch isn’t stupid, though. When he retires, he said, “I will probably install one.”

A lot of people in Moore have done just that since the 1999 tornado killed 43 people in the Oklahoma City area. Kitch says more than 10 percent of Moore’s homes — about 2,500 — now have a safe room or shelter. Helping homeowners make the investment: A federal program that has paid up to $2,000 of the cost.

Every person has to make decisions about what to spend on preparedness for natural disasters. Buy a house with a view, or live a few blocks from the beach? Build a storm shelter or make friends with a neighbor who has one? Get a weather radio? Buy earthquake insurance? Towns, too, must set building codes and choose whether to restrict the use of cheap building materials. They decide whether to allow development in flood plains, and whether to invest in sophisticated emergency equipment.

These are all down payments on the cost of a disaster. The rule of thumb is that every dollar spent on preparedness saves $4 in recovery costs, according to a report by the National Institute of Building Sciences’ Mutihazard Mitigation Council, which others confirm. That’s $4 taxpayers won’t have to spend.

In perhaps the closest thing to a bailout for those of us too small to matter, the government increasingly covers the costs of disasters.

The burden of disaster recovery is growing. In the 1950s, disasters in the United States caused a combined $53.6 billion in insured losses, according to an assessment by the Wharton Risk Management and Decision Processes Center at the University of Pennsylvania. In the 1990s, losses reached $778 billion.

Driving up the costs are population growth and urbanization. In 2004, Florida alone had $1.9 trillion in insured assets along its high-risk coastal areas, according to the Wharton Risk Management Center’s report.

The insurance industry is balking. Private insurers, stung by huge costs of doing business over the last decade, have stopped writing policies for some areas, or charge such high premiums that people decide to take their chances. Even the National Flood Insurance Program, which will run a $28 billion deficit after Hurricane Sandy, won’t insure against flooding in some areas.

Many assets aren’t insured, which places more of a burden on government recovery programs. Taxpayers paid out about 62 percent of the recovery costs of disasters between 2000 and 2008, the Wharton Risk Management center reported.

“You didn’t see the governor of New Jersey make a big deal about having only 30 to 40 percent market penetration for flood insurance,” after Hurricane Sandy, said Jeffrey Czajkowski, a research fellow at the Wharton center. He noted many people live in areas where they could get flood insurance from the government but do not. “The other 60 percent should have insurance, and if they don’t, why should they get fast aid?” he said.

The Federal Emergency Management Agency is the main source of disaster recovery funds. It’s also the primary source of money for preparedness, though groups like the Army Corps of Engineers also have budgets for preparedness projects.

In 2012, FEMA spent more than $7 billion on recovery. By comparison, it spent $2 billion on disaster preparedness for various events, including terrorism.

FEMA has drawn fire from Sen. Tom Coburn, R-Okla., for its slowness in paying out those funds; it has billions in unspent preparedness money on hand.

The agency keeps money in reserve for ongoing projects, but a high bar to get FEMA funds may also be part of the reason for the delay. Most of FEMA’s awards are reimbursement grants. In other words, a community or a person pays money for a project and then applies for reimbursement.

In the case of Salisbury, Mass., for example, the application process took “hundreds of hours over three years,” said Jerry Klima, a retired corporate lawyer and town selectman.

Salisbury got a grant to protect a business district and a portion of U.S. Route 1 from flooding that can close the highway for days at a time. The grant means the town can repair a culvert beneath an old rail bed that will drain water away from the commercial area during storms.

Politics can also bog down FEMA’s grants. In New Jersey, for example, some beachfront property owners are resisting community efforts to shore up protective dunes that would blunt the impact of coastal storms such as Sandy. Their lack of support threatens some plans, despite promised money from FEMA.

In a now-famous case, a couple won a judgment against their town for building a protective dune that blocked their ocean view — and eventually protected their property from Sandy’s impact.

 FEMA’s preparedness money “has been an issue of confusion over the years,” acknowledged Timothy W. Manning, deputy director for protection and national preparedness. The money is reserved for projects in process, such as transit station reinforcements or infrastructure improvements to ports, he said. FEMA has moved to speed its grant process, and now has $4.5 billion on hand, versus $9 billion a year ago.

FEMA’s former (and first) undersecretary of preparedness, George W. Foresman, said the agency has become tremendously more efficient since Hurricane Katrina hit the Gulf Coast in August 2005, but also said that “we get so focused on bells and whistles that we fail to remember that behind every successful program is a good accountant and a good accounting infrastructure.”

FEMA’s heightened emphasis on preparedness has had a broad effect, coinciding with a boost in preparedness by states and communities. A 2012 FEMA report shows that 69 percent of the population lives in the 18,000 communities with a disaster mitigation plan, up from 57 percent of the population in 2007.

More than 75 percent of communities are confident of their abilities to respond to a disaster, according to the FEMA report, and 95 percent have participated in preparation exercises.

FEMA uses money — and its build first, then rebate model — to induce local planning, said Louise Comfort at the University of Pittsburgh’s Center for Disaster Management.

“FEMA is trying to create some leverage on the community to do their hazard assessment, which all communities are supposed to do, and to identify the most critical hazards,” she said.

 The strategy of forcing communities to plan is working, she said. Insurer reluctance to write policies in areas identified as high-risk, such as those along the Gulf Coast, is causing people in those areas to move.

The insurance industry uses other incentives to coax communities to plan. More cities and towns are adopting the industry-sponsored Building Code Effectiveness Grading Schedule, which rates hazard preparedness from 1 to 10. (Only a handful of communities in the United States have the top rating of 1.) Better ratings mean lower insurance premiums.

Building requirements can make all the difference. In Moore, building codes didn’t change after the 1999 tornado. However, the head of a local developer, Home Creations Inc., decided he would never again build a house that didn’t include wind-resistant features such as specially designed J-bolts, tornado straps for the roof and sturdy, oriented-strand board across the entire house. (The state requires oriented-strand board only around windows and corners.) These features should hold up under 160 mph winds. They add about $3,000 to the cost of the house.

No amount of preparation will help a community avoid severe damage during an event like an EF-5 tornado. The one that hit Moore left a mile-wide path of destruction over a 50-mile long course, causing more than $1 billion in damage. It remains one of the costliest tornadoes on the books, though it was far surpassed by the monster EF-5 tornado that struck Joplin, Mo., in May 2011, wiping out a third of that city and causing an estimated $2.8 billion in damage.

In Moore, Kitch says the city, which lost more than 800 homes, spent about six weeks on initial cleanup. New homes were up within a year. A study three years later estimated the worst hit areas were 84 percent recovered.

Now, Kitch said, you wouldn’t even know those places had been leveled.

As for the future, he said, “We have tornadoes here in Oklahoma. That’s just a given.” And people have to know what they’re going to do when the next one hits.