Right now, even before the expected confirmation of another rate increase from the US Federal Reserve on Wednesday, going into the turn of the year, borrowing dollars is less affordable than previously.

In truth, many feel that the current year end squeeze in dollar funding costs is purely calendar-related. US firm BNY Mellon made the point last week that “financial institutions go to great lengths to ensure they have funding surplus-to-requirements over the festive season” though with the proviso that this year, “wrangling over the US debt ceiling as the year draws to a close” may be accentuating concerns.

But will a tide of affordable US dollar liquidity flow again in 2018? And if it does flow, will the incoming tide of greenbacks be priced as cheaply as before, or will the world’s banks find their funding costs are going to stay higher?

On Friday, Oliver Brennan, Macro Strategist at TS Lombard in London, wrote that the “year-end turn has clearly increased Europeans’ demand for dollars,” but cautioned that “there are structural factors which mean the squeeze can be painful and that the basis swap may not bounce back all the way.”

So a rising tide of US dollar liquidity might return in 2018, but access to that liquidity pool could be more expensive than previously.

One of the structural factors highlighted by TS Lombard is the introduction in January of rules under the Basel III regulations that urge banks to hold collateral “against [over-the-counter] transactions including FX swaps [which] will on average increase the cost of FX swaps and therefore widen the euro basis.”

And this isn’t just a problem for euro area banks.

Japanese banks, which have greatly expanded the size of their global footprint in US dollars in recent years in pursuit of better returns than are on offer in Japan, are also finding that the cost of borrowing greenbacks to finance their loan books has risen.

The Bank for International Settlements (BIS) noted in June that Japanese banks had “kept expanding both their gross and net US dollar positions, thereby creating substantial structural funding needs,” since the Global Financial Crisis.

More broadly, while Fitch Ratings “outlook for bank performance in Asia-Pacific is decidedly less negative than at the start of 2017” the potential for higher funding costs in 2018 is one of the pressures Fitch flags up.

Then there is the potential impact of higher US dollar funding costs on non-bank debt. The BIS has previously estimated the size of non-bank debt held by emerging market economies at US$11 trillion.

But why shouldn’t the tide of cheap US dollar liquidity rise again once end-of-year funding pressures have passed? Aside Basel III regulatory change, what else could disrupt the ecosystem of the liquidity pool on which the world’s banks rely?

The answers arguably lie in the United States in the confluence of developing US tax legislation, Federal Reserve monetary policy and the prospect of President Trump rolling out further economic stimulus via plans to encourage infrastructure investment.

US tax cuts will provide fiscal stimulus at a time when, certainly if the low level of US unemployment is a guide, the US economy doesn’t necessarily need such a helping hand. In the longer term, as the tax plan will add to the US national debt, Uncle Sam will be in the market to borrow more US dollars himself.

Against the backdrop of fiscal stimulus, and the separate possibility that proposals for upgrading US infrastructure may be unveiled in January, the Federal Reserve might feel a faster pace of interest rate increases is necessary at the same time as the US central bank continues to shrink its own balance sheet, taking greenbacks out of the system, through quantitative tightening.

Meanwhile, to the extent that US tax reforms could encourage US firms to repatriate US dollars from abroad it might, at the margin, push up the funding costs of banks outside the United States who rely on that offshore liquidity pool.

The world may have to get re-accustomed in 2018 to higher borrowing costs for US dollars.