This month, Washington will reach the $12.1 trillion statutory ceiling on the national debt. It will mark the 91st time Congress has had to raise the debt limit since the original $43 billion ceiling was established in 1940.

Of course, Congress cannot responsibly refuse to raise the debt limit: Doing so would prevent Washington from paying for past commitments, and would certainly cost the federal government its AAA bond rating. But this increase is for past fiscal profligacy and does nothing to impose future discipline.

The key question is: Will Congress attach real budget reforms to this vote? Or will it choose to retain the same failed budgeting practices that will guarantee a 92nd debt limit increase sometime next year? Real budget reform is necessary, and it must include statutory spending caps.

President Obama has proposed a budget that would double the national debt over the next decade, dumping $100,000 per household of additional debt into the laps of our children and grandchildren -- the equivalent of being handed a new mortgage, but without the house. This is not only economically unacceptable, it is also morally unacceptable.

Closing this fiscal gap with tax increases would cost $10,000 per household annually. Yet it's not clear why drowning future generations in exorbitant taxes is any better than drowning them in debt. Such large tax increases would also devastate the economy, leading to slower growth, fewer jobs and lower incomes. Besides, historic spending increases -- not declining tax revenues -- are responsible for this enlarged fiscal gap. By 2019, tax revenues are projected to approach the historical average of 18 percent of the economy. But Mr. Obama would push spending (historically 20 percent of the economy) to more than 28 percent.

Restraining spending is the only way to avoid the painful choice between massive budget deficits and painful tax increases. And statutory spending caps are an essential ingredient to spending restraint. In the current absence of spending caps, lawmakers can simply add up the cost of their preferred programs and pass legislation to fund them without setting priorities or making trade-offs.

Entitlement programs grow each year without any cap or oversight. Discretionary programs grow with only slight constraints imposed by the annual budget resolutions, which are routinely bypassed through the "emergency" designation. Not surprisingly, spending has expanded 51 percent faster than inflation since 2001.

At minimum, lawmakers should demand that any debt-limit increase also statutorily cap discretionary spending growth at the inflation rate (approximately 2.5 percent annually) for the next decade. This growth rate matches the successful discretionary caps of the 1990s, and also the president's current discretionary spending estimates from 2011 through 2019. A two-thirds supermajority requirement to bypass the spending caps -- even during emergencies -- can close the emergency loophole while assuring that any true emergency will still likely clear that threshold.

While capping discretionary spending would be helpful, an even better approach would cap total federal spending growth, including entitlements. After all, there is no reason for Washington to essentially keep entitlement programs (60 percent of all federal spending) on autopilot, growing 6 to 8 percent annually without oversight. Government-wide spending caps would force Congress to set priorities and make trade-offs among all federal programs. No more blank checks. Large spending increases for high-priority programs would require offsets from lower-priority programs.

Ideally, government-wide spending caps would set a simple and predictable annual growth rate. Tying spending growth to tax revenues -- which have recently fluctuated 15 percent annually -- would be too volatile. Tying spending growth to nominal economic growth (the effective result of current proposals to cap spending at 20 percent of the economy) would likely also be too volatile. In that situation, lawmakers would happily raise spending by 6 and 7 percent when an economic boom allows it, but quickly bypass required spending cuts during a recession. That's exactly what happened when lawmakers recently tried to tie yearly highway spending to gas-tax receipts.

One solution would tie overall spending growth to the previous five-year average of economic growth rates (which would smooth out the year-to-year extremes), so current proposals to cap spending at 20 percent of the economy would be more easily attained. Another option would cap spending at a more predictable target such as the inflation rate plus population growth (typically 3 to 5 percent annually), or even a pre-set rate such as 4 percent annually. This predictability would increase the likelihood of congressional compliance.

Spending caps are a necessary constraint to help -- or force -- Congress to close the budget deficit. Lawmakers could likely balance the budget by 2019 without tax increases by returning total federal spending to the 20 percent of the economy level that prevailed before the recession, and then capping its growth at inflation plus population growth. Working within this cap would not be easy, and it would require reforming Social Security, Medicare and Medicaid.

That's why the debt-limit increase should also include a bipartisan entitlement reform commission, whose recommendations would be guaranteed a congressional vote. However, the only alternative is choosing between $10,000-per-household tax increases and a doubling of the national debt. Surely, Congress can do better.

Brian M. Riedl is Grover M. Hermann Fellow in Federal Budgetary Affairs in the Thomas A. Roe Institute for Economic Policy Studies at The Heritage Foundation.