Within Mesoamerica, at least for the Maya area, labor seems to have been a sort of compulsory tax to a local lord or king. This doesn't mean that everyone had to provide work every year or even every other year, but when called upon they would be expected to provide labor for a project. The people probably went unpaid for their service to these projects since they benefited from services provided by the state such as protection from other cities and being able to take advantage of goods through trade routes.

These projects were usually built during the dry season which is advantageous for a couple of reasons. The first is that without crops to tend a population has somewhat more free time in which they could devote to labor projects. The second is that without the heavy rains of the rainy season a lot of construction projects were much easier to build. Things like pyramids, sacbeobs (roads), and palaces made use of a lot of earthen or clay fill which was then capped with cut stone or cobbles set in a sort of cement or mortar. Finally things were covered with a heavy layer of plaster which allowed one to decorate the building, but also further provided protection from the elements. This plaster would, of course, need to be reapplied from time to time as the rains from the rainy season took their toll.

That being said, most of the construction material in Mesoamerica came from close by, including the rocks and stones to build walls. In the Maya region, where many walls were constructed during this turbulent time period, the walls were constructed from collected limestone cobbles on the surface. Sometimes they would be roughly shaped to fit better together. Besides the later city of Mayapan, a lot of these walls seem to have been hastily built around the ceremonial centers of the city which were normally unprotected in such a manner but not necessarily "open" to anyone. Some of these Maya sites show that they did indeed make use of these walls in defense and other sites appeared to have been abandoned before any of the fighting reached them.

Mayapan, which is a much later city, was almost completely walled off during its initial construction with settlements continuing to spread outward beyond the wall's protection as the city expanded and grew. As you can see in this high density LiDAR map of the site, the area is densly populated with each of those mounds representing a pyramid, palace, or housing complex.

While I'm not aware of any energetics analyses on the walls of Mayapan or any other walled site, analyses have been done on residential structures at the site of Copan in Honduras. The palace for the 13th ruler, for example, required roughly 300 people working 83 days to completely build the structure during a single construction event even though the reality is that the palace was built in stages over time. So one can imagine, the walls of Mayapan most likely took a much larger amount of people a much longer amount of time to construct even if the walls were just collected cobbles from the surface.

I can offer some insight as regards Anglo-Saxon England (talking roughly 600 to 1066 AD).

Beginning in the eighth century, Anglo-Saxon charters began to include what has come to be known as the three obligations (or some other variant of that name). Those three are bridge-building, fortification-building, and military service. The obligations were extended even to religious institutions, and are described well in this excerpt of a charter from Offa to Abbot Stithberht:

they [the lands granted] should be free in perpetuity, value for value, from the payment of all royal taxes, works and burdens, and also the exactions of popular assemblies, except only an account must be rendered for the three public burdens, that is, the building of bridges and fortress, but also now only 5 men should be sent for the requirements of military service.

In the grand scheme, these reservations mark the beginning of a change from charters of the seventh century to the more stipulation-heavy charters of the ninth through eleventh centuries. A charter from Hlothere, King of Kent, in the late seventh century, is similar in character to the above-mentioned charter of Offa; a bequeath to an abbot, a fairly sizeable piece of land, but the essence of the land grant is as follows:

I, Hlothhere, king of the dwellers in Kent, for the salvation of my soul give the land in Thanet that is called Westana to you, Beorhtwald and to your monastery with all that belongs to it, fields, pastures, marshes, small woods, streams, fisheries, all, as has been said, belonging to this same land, just as has ever been possessed up until now

There is no mention of the three obligations, nor of any other exemptions Offa gave, from royal taxes etcetera. From the late eighth century onwards, these stipulations become the norm, eventually being condensed down to a short phrase,

Moreover, may the aforementioned land be free from all yoke of earthly servitude, with three exceptions, namely the fixed military service [and] the restoration of bridge and fortress.

which is roughly consistent in charters across the tenth century (above from a charter of King Edgar to the church of St. Peter at Bath).

It is a point of contention, whether these obligations were added around the reign of Offa, or whether they were implied beforehand and the late eighth century simply marks the beginning of their written transcription. At any rate, in response to your question, what the charters describe is the king's reservation of fortress construction as his right to demand. This was likely manifest in labour-rents, no new idea in Anglo-Saxon England. Labour on a lord's farmland was an expectation of peasants, especially those living on a lord's inland. These inland peasants were bound tenants, and part of their obligations was labour. The eleventh century Rectitudines Singularum Personarum outlines a variety of work schemes for different ranks of people, anywhere from one to three days per week as well as other obligations like reaping hay-meadows, maintaining enclosures, etcetera.

What seems to be going on with fortress construction, then, is just a part of the obligations due to a lord (in this case the king). It is possible this obligation was used to construct Offa's Dyke. David Hill has pointed out that there is a large amount of variability from one segment of the dyke to the next, perhaps indicating highly localised musters of a workforce, who would construct the section nearest them. As far as actual town defences go, Jeremy Haslam has analysed the defensive structures around Daws Castle in Wessex, and has likewise concluded that the earliest phase of construction - somewhere between the late eighth and mid-ninth centuries - was likely sourced from the obligation of fortress building (which appeared in writing around the mid eighth century in Wessex). Haslam notes, and I agree, that rising Viking pressure may have put an impetus on these early constructions, and perhaps on the emergence of fortress-building in formal writing.

As part of a general obligation, these projects probably took place without a fundraising effort or exchange of money like you mentioned. Rather, the labour-rent was just a facet of the overall dues peasants owed their lords.

Sources:

"Rectitudines Singularum Personarum", ed. Thom Gobbitt, from Early English Laws

For charters, the Electronic Sawyer, S106, 8, and 777 (in the order of their appearance in my answer)

"The Construction of Offa's Dyke," David Hill, Antiquaries Journal 65, issue 1

"Daws Castle, Somerset, and Civil Defence Measuresin Southern and Midland England in the Ninth to Eleventh Centuries," Jeremy Haslam, Archaeological Journal 168, issue 1