UPS Raises Rates 5.9% From 21 December as US Q4 Parcel Costs Peak

The Analysis

UPS has told the market it will take published rates up 5.9% from 21 December, and every headline I have seen leads with the number. I lead with the date. The twenty-first of December sits four days out from Christmas, inside the last full delivery week, and past the point where anything you hand over still has a realistic path to a doorstep before the twenty-fifth. A general rate increase landing on that date is not really a December event. It is a January event wearing a December costume, and if you budget for it as a December cost you will put your money in the wrong week.

Here is what December actually looks like from where I sit. The money that hurts this quarter was switched on in October. Peak and demand surcharges, residential delivery fees, remote area fees, oversize and additional handling charges: those went live weeks ago and they are billed per package, on top of whatever the base rate happens to be. By the time the 5.9% shows up on an invoice, most December parcels will already have been billed, delivered, or lost. The increase matters for the January replenishment wave, for the returns wave behind it, and for whoever is still tendering hard between the twenty-first and the last pre-holiday cut-off, where a new base rate and a peak table can overlap.

Two other things landed in the same news cycle and they belong in the same conversation. Amazon has pushed the United States Black Friday and Cyber Monday deal submission window out to 13 November, which sounds generous until you count backwards from a 21 December rate date and a mid-December fulfilment cut-off. Amazon has also made a dangerous goods compliance questionnaire mandatory across four categories. The first one moves your calendar. The second one moves your queue, and the queue is the part that can cost you a season.

Start with the two numbers nobody is disputing. UPS goes up 5.9% on 21 December. Industry trackers expect peak season shipping cost per package in the United States to set a record. Those two sentences get printed next to each other as if one explains the other, and they do not. A general rate increase is a list price action, decided in a planning cycle, applied to a base rate. Cost per package is an outcome, and it is an outcome dominated by accessorials rather than by the base. When trackers say cost per package will set a record, the thing setting that record is the surcharge stack that has been building since October plus a mix shift towards residential and remote deliveries that nobody can reverse in November. Confusing the two is how people end up negotiating the wrong line on the invoice.

Why now. Carriers worked out years ago that an increase announced in the middle of peak reads as greed and generates phone calls, while one announced for the tail of peak reads as routine and generates shrugs. Putting it on 21 December does something more useful for them: it lands after the most price sensitive shipping days of the year and before the January contract conversations, so the increase is already baked in when you sit down to talk about next year. Read that as a calendar tell. The date was not chosen to hit your December profit line. It was chosen to set the anchor for your January negotiation, and the accessory benefit is that almost nobody notices it arrive.

Who eats it, and how much. Split your own book three ways and the answer changes completely. Low average order value parcels get hit hardest as a percentage. A fifteen dollar order in a two pound box does not care about a 5.9% move on a nine dollar base rate, because that is fifty-three cents. It cares very much about a peak surcharge of a dollar seventy-five, a residential fee of sixty cents, and an additional handling or oversize fee if the box falls on the wrong side of a dimension threshold. On that parcel the accessorials run three to five times the increase. Shippers with two hundred dollar baskets see the same accessorials and barely notice them, which is why one headline produces panic in one internal channel and silence in another.

Remote and extended area delivery is the sleeper, and it is the line I would look at before any other. A parcel going to a rural postcode can pick up a surcharge several times the size of the general increase, and it lands on exactly the customers who already wait longest for a truck. If six percent of your December volume sits in those codes, that single line can beat the whole 5.9% across your entire book. Pull your own postcode distribution for last December before you argue about rates with anybody. Case-packed replenishment heading to a distribution centre sits on a different curve entirely: none of this is your story, and your actual problem is appointment capacity at the receiving dock in the first week of January, when everyone's returns and everyone's replenishment turn up on the same morning.

Now the part I get paid to think about, which is which hour breaks first. Walk one parcel from the moment a customer clicks to the moment it is scanned at a door. The first failure point is not transit. It is the tender window. Every origin facility has a cut-off, typically late afternoon, and a parcel that misses it does not lose an hour, it loses a day, because it waits for the next outbound linehaul. In a normal week that costs you twenty-four hours. In the week before Christmas it costs you the holiday, because the next outbound linehaul is already committed to parcels that made the previous day's cut-off and the network is sorting to capacity overnight. Miss a five in the afternoon tender on the eighteenth and you are not arriving on the twenty-third. You are arriving on the twenty-seventh, at a house where nobody is home and the present has already been replaced.

The next failure point is the injection scan, and this is the one nobody puts on a customer facing page. A label printed but not scanned is a parcel that does not exist. On a normal day the gap between label creation and first physical scan is a few hours. In the last ten days of December that gap stretches, and it stretches unevenly, because a parcel collected by a driver at a busy branch can sit sixteen to thirty hours before anyone scans it into the network. If your customer service team quotes transit times off label creation, every promise made inside that gap is already broken and you do not know it yet. Move your promise clock to first physical scan and you stop lying to people by accident.

After that comes the sort. Overnight sorts run roughly from ten at night to five in the morning, and the peak of the peak is the two to four window, when the plant is running at the top of its design throughput and any exception parcel, a bad address, a dim weight dispute, a damaged label, gets pulled onto a manual line that is already an hour deep. That is where a December parcel goes to die. Not on a truck, not on an aircraft. On a conveyor belt at three in the morning with a human being squinting at a smudged label.

Then the last mile, and here is the customer promise arithmetic. If your site says order by 20 December for Christmas delivery, it is saying something that was true in October and is not true in December. Working backwards from delivery on the twenty-fifth, with a weekend in the way and one day of buffer for a missed scan, you want the parcel injected no later than the evening of the seventeenth and tendered no later than the afternoon of the sixteenth. That is three to four days earlier than most websites currently say, and the gap between saying it and not saying it is the gap between a support queue and a chargeback queue in January.

Here is the thing I have not seen in any of the coverage, and it is the reason I am writing this rather than forwarding the press release. Everybody is doing arithmetic on 5.9%. Nobody is doing arithmetic on the fact that the increase takes effect inside a window where demand surcharges are still running. Assume this season's peak table follows the pattern of the last couple of seasons and runs into the middle of January rather than ending at Christmas. Then the stretch from 21 December to roughly mid-January is an overlap period: new base rate and peak accessorials at the same time, on every parcel. That is not a 5.9% event. On a cheap parcel it is closer to a fifteen to twenty percent event, because accessorials are flat per package amounts and they do not shrink just because the base moved.

And look at what moves in that window. It is not the gift parcels. It is returns going back, and replenishment going out to restock whatever sold, and both of those move at the thinnest margin of the year. The increase was announced for a date inside peak, which means its first full month of real effect lands on the least profitable parcels you ship all year, right when your finance team has stopped looking at freight lines because the season is over. If you want one sentence to take into a budget meeting, take that one.

There is a second version of the same mistake hiding in the Amazon news. Extending deal submission to 13 November reads like extra time, and it is extra time to submit, not extra time to be approved. A mandatory dangerous goods questionnaire across four categories is a review step sitting between submission and a live deal. If your catalogue touches anything with a lithium cell, an aerosol, a solvent, a fragrance, a pressurised container, or a magnet that raises an eyebrow, each affected listing has to clear that review, and reviews clear at the pace of the queue rather than at the pace of your calendar. Submitting on the thirteenth is easy.

Being approved by the fulfilment inbound cut-off is a different question, and the only way to know the answer is to have submitted weeks ago. Anyone treating 13 November as the deadline has already spent the buffer they were just handed. There is also the inventory side of the same trap: a deal approved with nothing received into the building is a listing that goes live and then goes out of stock, which burns rank and burns the ad spend behind it. The submission window only matters if the goods are physically in place before the cut-off, and that clock is set by your booking date, not by Amazon's form.

Let me put numbers on it, and I will state every assumption, because half the arguments people have in December are two people comparing different baskets. Assume a two pound residential parcel in zone five with a base rate of ten dollars before the increase. Assume a peak demand surcharge of one dollar seventy-five per package, live since October. Assume a residential delivery fee of sixty cents and an extended area fee of five dollars sixty on six percent of volume. Assume ten thousand packages in December, of which four thousand are tendered on or after 21 December. Assume a further eight hundred packages in the final rush needing expedited handling at eight dollars each.

The general increase contributes four thousand packages times fifty-nine cents, which is two thousand three hundred and sixty dollars. The peak surcharge contributes ten thousand times one dollar seventy-five, seventeen thousand five hundred. Residential contributes six thousand. Extended area on six hundred packages contributes three thousand three hundred and sixty. Expedited handling on eight hundred packages contributes six thousand four hundred. Total accessorial and increase money on that book is thirty-five thousand six hundred and twenty dollars. The 5.9% is two thousand three hundred and sixty of it, which is about six point six percent.

Said the other way round, accessorials are running at roughly fifteen times the general increase on the same book, averaging three dollars fifty-six per package across the whole ten thousand against fifty-nine cents on the four thousand the increase actually touched. On a fifteen dollar order, three dollars fifty-six is the difference between a contribution margin and a loss. If you spend your week negotiating the 5.9% and leave the other ninety-three percent alone, you have done the easy six percent of the work.

One lever gets ignored every year because it is boring, and it is the only lever you can still pull in November. Dimensional weight. A two pound parcel in a box that is two inches too tall can bill at a weight class above its actual weight, and the divisor is applied before every accessorial on the invoice, which means the surcharge stack rides on the inflated number too. Go through your top twenty stock keeping units by volume this week with a tape measure and a scale, and look at what happens to the bill if each one drops a size. Right-sizing packaging is not a sustainability project, it is a rate negotiation you win by yourself, and it is the only cost line in this whole piece that you control end to end.

The support desk is the other place December is won or lost, and almost nobody staffs it against the carrier clock. If your network cuts off at five in the afternoon and your support team signs off at five, then every promise made after five was made into a void. Either move the last order cut-off on your own site to ninety minutes before the carrier cut-off, or staff the desk to cover the gap, and put the real last-tender time on the same page as the delivery promise so that the person answering the chat and the person printing labels are reading the same number. I have watched a December go sideways purely because the website said one thing and the branch did another, and the customer does not care which of you was wrong.

Now the triage question, because this is the one that decides whether December is survivable at all. When capacity tightens in the last ten days you cannot ship everything, so you pick, and the rule I use is revenue at risk per hour of network capacity. A two hundred dollar order for a repeat customer with a stated delivery date consumes the same linehaul slot as a fifteen dollar order for a first time buyer. If you can only move one, the answer is not whichever entered the queue first. Build the triage list in the first week of December rather than on the nineteenth: top decile customers by lifetime value with a delivery promise still open, then anything with a hard external date, then everything else. Tell the bottom of the list early. A customer told on the twelfth that delivery will slip to the twenty-eighth is annoyed. A customer told on the twenty-sixth that it has slipped is gone, and they take their January with them.

Where this could be wrong. The 5.9% is a list action. If you are on a negotiated agreement with an increase cap, an index linked cap, or a fixed percentage ceiling, your effective increase may be materially lower and it is worth reading that clause before you accept the number. If December volumes come in under what you committed, your leverage on the accessorial side improves sharply in January, because carriers price accessorial relief the same way they price volume.

Two things would flip the conclusion entirely. If average order value across your book is above roughly a hundred dollars, accessorials stop being the story and service failure becomes the story, because a parcel arriving on the twenty-seventh is a refund and a lost customer whatever it cost to ship. If your mix is mostly commercial addresses, the residential and extended area lines shrink and 5.9% genuinely does become your largest single line, at which point negotiating the increase is exactly the right use of your week.

The other real branch is carrier diversification. Regional carriers and the postal network do not carry the same peak accessorial structure, and for volume inside a metro region they can beat a national carrier in December on both price and cut-off. The catch is coverage: hand rural and remote volume to a regional partner and you have swapped a surcharge problem for a service problem, and you will not find out until the second week of January, when the complaints arrive in one batch.

So what do you actually do this week, given that October is nearly over and the 13 November window is a fortnight away. Stop treating the rate increase as the December problem. Treat the tender clock, the injection scan and the accessorial expiry dates as the December problem, and treat the 5.9% as what it really is, which is the opening number of your January negotiation.

↑ Back to the key points

— By Nadia Pryce